+ close()
Close httplib2 connections.
+ +Gets details of a single Agent under an AI Application.
+
+ list(parent, pageSize=None, pageToken=None, x__xgafv=None)
Lists Agents under an AI Application.
+
+ list_next(previous_request, previous_response)
Retrieves the next page of results.
+close()
+ Close httplib2 connections.+
get(name, x__xgafv=None)
+ Gets details of a single Agent under an AI Application.
+
+Args:
+ name: string, Required. Name of the resource. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}/agents/{agent}` (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Represents an Agent. "A2A" below refers to the Agent-to-Agent protocol.
+ "agentId": "A String", # Output only. A stable, globally unique identifier for agents.
+ "attributes": { # Output only. Attributes of the Agent. Valid values: * `agentregistry.googleapis.com/system/Framework`: {"framework": "google-adk"} - the agent framework used to develop the Agent. Example values: "google-adk", "langchain", "custom". * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the Agent. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Agent, for example, the Reasoning Engine URI.
+ "a_key": {
+ "a_key": "", # Properties of the object.
+ },
+ },
+ "card": { # Full Agent Card payload, often obtained from the A2A Agent Card. # Output only. Full Agent Card payload, when available.
+ "content": { # Output only. The content of the agent card.
+ "a_key": "", # Properties of the object.
+ },
+ "type": "A String", # Output only. The type of agent card.
+ },
+ "createTime": "A String", # Output only. Create time.
+ "description": "A String", # Output only. The description of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no description.
+ "displayName": "A String", # Output only. The display name of the agent, often obtained from the A2A Agent Card.
+ "location": "A String", # Output only. The location where agent is hosted. The value is defined by the hosting environment (i.e. cloud provider).
+ "name": "A String", # Identifier. The resource name of an Agent. Format: `projects/{project}/locations/{location}/agents/{agent}`.
+ "protocols": [ # Output only. The connection details for the Agent.
+ { # Represents the protocol of an Agent.
+ "interfaces": [ # Output only. The connection details for the Agent.
+ { # Represents the connection details for an Agent or MCP Server.
+ "protocolBinding": "A String", # Required. The protocol binding of the interface.
+ "url": "A String", # Required. The destination URL.
+ },
+ ],
+ "protocolVersion": "A String", # Output only. The version of the protocol, for example, the A2A Agent Card version.
+ "type": "A String", # Output only. The type of the protocol.
+ },
+ ],
+ "skills": [ # Output only. Skills the agent possesses, often obtained from the A2A Agent Card.
+ { # Represents the skills of an Agent.
+ "description": "A String", # Output only. A more detailed description of the skill.
+ "examples": [ # Output only. Example prompts or scenarios this skill can handle.
+ "A String",
+ ],
+ "id": "A String", # Output only. A unique identifier for the agent's skill.
+ "name": "A String", # Output only. A human-readable name for the agent's skill.
+ "tags": [ # Output only. Keywords describing the skill.
+ "A String",
+ ],
+ },
+ ],
+ "uid": "A String", # Output only. A universally unique identifier for the Agent.
+ "updateTime": "A String", # Output only. Update time.
+ "version": "A String", # Output only. The version of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no version or agent is not an A2A Agent.
+}
+list(parent, pageSize=None, pageToken=None, x__xgafv=None)
+ Lists Agents under an AI Application.
+
+Args:
+ parent: string, Required. Parent value (AI Application) for ListAiApplicationAgentsRequest. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required)
+ pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default.
+ pageToken: string, Optional. A token identifying a page of results the server should return.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Message for response to listing Agents under an AI Application.
+ "agents": [ # The list of Agents.
+ { # Represents an Agent. "A2A" below refers to the Agent-to-Agent protocol.
+ "agentId": "A String", # Output only. A stable, globally unique identifier for agents.
+ "attributes": { # Output only. Attributes of the Agent. Valid values: * `agentregistry.googleapis.com/system/Framework`: {"framework": "google-adk"} - the agent framework used to develop the Agent. Example values: "google-adk", "langchain", "custom". * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the Agent. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Agent, for example, the Reasoning Engine URI.
+ "a_key": {
+ "a_key": "", # Properties of the object.
+ },
+ },
+ "card": { # Full Agent Card payload, often obtained from the A2A Agent Card. # Output only. Full Agent Card payload, when available.
+ "content": { # Output only. The content of the agent card.
+ "a_key": "", # Properties of the object.
+ },
+ "type": "A String", # Output only. The type of agent card.
+ },
+ "createTime": "A String", # Output only. Create time.
+ "description": "A String", # Output only. The description of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no description.
+ "displayName": "A String", # Output only. The display name of the agent, often obtained from the A2A Agent Card.
+ "location": "A String", # Output only. The location where agent is hosted. The value is defined by the hosting environment (i.e. cloud provider).
+ "name": "A String", # Identifier. The resource name of an Agent. Format: `projects/{project}/locations/{location}/agents/{agent}`.
+ "protocols": [ # Output only. The connection details for the Agent.
+ { # Represents the protocol of an Agent.
+ "interfaces": [ # Output only. The connection details for the Agent.
+ { # Represents the connection details for an Agent or MCP Server.
+ "protocolBinding": "A String", # Required. The protocol binding of the interface.
+ "url": "A String", # Required. The destination URL.
+ },
+ ],
+ "protocolVersion": "A String", # Output only. The version of the protocol, for example, the A2A Agent Card version.
+ "type": "A String", # Output only. The type of the protocol.
+ },
+ ],
+ "skills": [ # Output only. Skills the agent possesses, often obtained from the A2A Agent Card.
+ { # Represents the skills of an Agent.
+ "description": "A String", # Output only. A more detailed description of the skill.
+ "examples": [ # Output only. Example prompts or scenarios this skill can handle.
+ "A String",
+ ],
+ "id": "A String", # Output only. A unique identifier for the agent's skill.
+ "name": "A String", # Output only. A human-readable name for the agent's skill.
+ "tags": [ # Output only. Keywords describing the skill.
+ "A String",
+ ],
+ },
+ ],
+ "uid": "A String", # Output only. A universally unique identifier for the Agent.
+ "updateTime": "A String", # Output only. Update time.
+ "version": "A String", # Output only. The version of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no version or agent is not an A2A Agent.
+ },
+ ],
+ "nextPageToken": "A String", # A token identifying a page of results the server should return.
+}
+list_next(previous_request, previous_response)
+ Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++
+ close()
Close httplib2 connections.
+ +Gets details of a single Endpoint under an AI Application.
+
+ list(parent, pageSize=None, pageToken=None, x__xgafv=None)
Lists Endpoints under an AI Application.
+
+ list_next(previous_request, previous_response)
Retrieves the next page of results.
+close()
+ Close httplib2 connections.+
get(name, x__xgafv=None)
+ Gets details of a single Endpoint under an AI Application.
+
+Args:
+ name: string, Required. Name of the resource. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}/endpoints/{endpoint}` (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Represents an Endpoint.
+ "attributes": { # Output only. Attributes of the Endpoint. Valid values: * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Endpoint, for example, the GKE Deployment.
+ "a_key": {
+ "a_key": "", # Properties of the object.
+ },
+ },
+ "createTime": "A String", # Output only. Create time.
+ "description": "A String", # Output only. Description of an Endpoint.
+ "displayName": "A String", # Output only. Display name for the Endpoint.
+ "endpointId": "A String", # Output only. A stable, globally unique identifier for Endpoint.
+ "interfaces": [ # Required. The connection details for the Endpoint.
+ { # Represents the connection details for an Agent or MCP Server.
+ "protocolBinding": "A String", # Required. The protocol binding of the interface.
+ "url": "A String", # Required. The destination URL.
+ },
+ ],
+ "name": "A String", # Identifier. The resource name of the Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "updateTime": "A String", # Output only. Update time.
+}
+list(parent, pageSize=None, pageToken=None, x__xgafv=None)
+ Lists Endpoints under an AI Application.
+
+Args:
+ parent: string, Required. Parent value (AI Application) for ListAiApplicationEndpointsRequest. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required)
+ pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default.
+ pageToken: string, Optional. A token identifying a page of results the server should return.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Message for response to listing Endpoints under an AI Application.
+ "endpoints": [ # The list of Endpoints.
+ { # Represents an Endpoint.
+ "attributes": { # Output only. Attributes of the Endpoint. Valid values: * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Endpoint, for example, the GKE Deployment.
+ "a_key": {
+ "a_key": "", # Properties of the object.
+ },
+ },
+ "createTime": "A String", # Output only. Create time.
+ "description": "A String", # Output only. Description of an Endpoint.
+ "displayName": "A String", # Output only. Display name for the Endpoint.
+ "endpointId": "A String", # Output only. A stable, globally unique identifier for Endpoint.
+ "interfaces": [ # Required. The connection details for the Endpoint.
+ { # Represents the connection details for an Agent or MCP Server.
+ "protocolBinding": "A String", # Required. The protocol binding of the interface.
+ "url": "A String", # Required. The destination URL.
+ },
+ ],
+ "name": "A String", # Identifier. The resource name of the Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "updateTime": "A String", # Output only. Update time.
+ },
+ ],
+ "nextPageToken": "A String", # A token identifying a page of results the server should return.
+}
+list_next(previous_request, previous_response)
+ Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++
+ agents()
+
Returns the agents Resource.
+ +
+ endpoints()
+
Returns the endpoints Resource.
+ +
+ mcpServers()
+
Returns the mcpServers Resource.
+Close httplib2 connections.
+ +Gets details of a single AI Application.
getIamPolicy(resource, options_requestedPolicyVersion=None, x__xgafv=None)
Gets the access control policy for a resource. Returns an empty policy if the resource exists and does not have a policy set.
+
+ list(parent, filter=None, orderBy=None, pageSize=None, pageToken=None, x__xgafv=None)
Lists AI Applications in a given project and location.
+
+ list_next(previous_request, previous_response)
Retrieves the next page of results.
setIamPolicy(resource, body=None, x__xgafv=None)
Sets the access control policy on the specified resource. Replaces any existing policy. Can return `NOT_FOUND`, `INVALID_ARGUMENT`, and `PERMISSION_DENIED` errors.
@@ -92,6 +116,66 @@Close httplib2 connections.+
get(name, x__xgafv=None)
+ Gets details of a single AI Application.
+
+Args:
+ name: string, Required. Target AI Application resource name. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Represents an AI Application.
+ "applicationProperties": { # Additional system properties of an Application. # Output only. Properties of an underlying cloud resource that can comprise an AI Application.
+ "extendedMetadata": { # Output only. Additional metadata specific to the App Hub application. The key is a string that identifies the type of metadata and the value is the metadata contents specific to that type. Key format: `apphub.googleapis.com/{metadataType}`
+ "a_key": { # Additional metadata for a Service or Workload.
+ "metadataStruct": { # Output only. The metadata contents.
+ "a_key": "", # Properties of the object.
+ },
+ },
+ },
+ },
+ "attributes": { # Consumer provided attributes. # Optional. Consumer provided attributes.
+ "businessOwners": [ # Optional. Business team that ensures user needs are met and value is delivered
+ { # Contact information of stakeholders.
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "criticality": { # Criticality of the Application, Service, or Workload # Optional. User-defined criticality information.
+ "type": "A String", # Required. Criticality Type.
+ },
+ "developerOwners": [ # Optional. Developer team that owns development and coding.
+ { # Contact information of stakeholders.
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "environment": { # Environment of the Application, Service, or Workload # Optional. User-defined environment information.
+ "type": "A String", # Required. Environment Type.
+ },
+ "operatorOwners": [ # Optional. Operator team that ensures runtime and operations.
+ { # Contact information of stakeholders.
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ },
+ "createTime": "A String", # Output only. Creation time.
+ "description": "A String", # Optional. User-defined description of the AI Application.
+ "displayName": "A String", # Optional. User-defined name for the AI Application.
+ "name": "A String", # Identifier. Resource name of the AI Application. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}`
+ "state": "A String", # Output only. The current state of the AI Application.
+ "uid": "A String", # Output only. Universally unique identifier (UUID4) for the AI Application.
+ "updateTime": "A String", # Output only. Last update time.
+}
+getIamPolicy(resource, options_requestedPolicyVersion=None, x__xgafv=None)
Gets the access control policy for a resource. Returns an empty policy if the resource exists and does not have a policy set. @@ -140,6 +224,92 @@Method Details
}
list(parent, filter=None, orderBy=None, pageSize=None, pageToken=None, x__xgafv=None)
+ Lists AI Applications in a given project and location.
+
+Args:
+ parent: string, Required. Parent project and location to query. Format: `projects/{project}/locations/{location}` (required)
+ filter: string, Optional. Filtering results.
+ orderBy: string, Optional. Hint for how to order the results.
+ pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default.
+ pageToken: string, Optional. A token identifying a page of results the server should return.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Message for response to listing AiApplications.
+ "aiApplications": [ # The list of AiApplications.
+ { # Represents an AI Application.
+ "applicationProperties": { # Additional system properties of an Application. # Output only. Properties of an underlying cloud resource that can comprise an AI Application.
+ "extendedMetadata": { # Output only. Additional metadata specific to the App Hub application. The key is a string that identifies the type of metadata and the value is the metadata contents specific to that type. Key format: `apphub.googleapis.com/{metadataType}`
+ "a_key": { # Additional metadata for a Service or Workload.
+ "metadataStruct": { # Output only. The metadata contents.
+ "a_key": "", # Properties of the object.
+ },
+ },
+ },
+ },
+ "attributes": { # Consumer provided attributes. # Optional. Consumer provided attributes.
+ "businessOwners": [ # Optional. Business team that ensures user needs are met and value is delivered
+ { # Contact information of stakeholders.
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "criticality": { # Criticality of the Application, Service, or Workload # Optional. User-defined criticality information.
+ "type": "A String", # Required. Criticality Type.
+ },
+ "developerOwners": [ # Optional. Developer team that owns development and coding.
+ { # Contact information of stakeholders.
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "environment": { # Environment of the Application, Service, or Workload # Optional. User-defined environment information.
+ "type": "A String", # Required. Environment Type.
+ },
+ "operatorOwners": [ # Optional. Operator team that ensures runtime and operations.
+ { # Contact information of stakeholders.
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ },
+ "createTime": "A String", # Output only. Creation time.
+ "description": "A String", # Optional. User-defined description of the AI Application.
+ "displayName": "A String", # Optional. User-defined name for the AI Application.
+ "name": "A String", # Identifier. Resource name of the AI Application. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}`
+ "state": "A String", # Output only. The current state of the AI Application.
+ "uid": "A String", # Output only. Universally unique identifier (UUID4) for the AI Application.
+ "updateTime": "A String", # Output only. Last update time.
+ },
+ ],
+ "nextPageToken": "A String", # A token identifying a page of results the server should return.
+ "unreachable": [ # Unordered list. Locations that could not be reached.
+ "A String",
+ ],
+}
+list_next(previous_request, previous_response)
+ Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++
setIamPolicy(resource, body=None, x__xgafv=None)
Sets the access control policy on the specified resource. Replaces any existing policy. Can return `NOT_FOUND`, `INVALID_ARGUMENT`, and `PERMISSION_DENIED` errors. diff --git a/docs/dyn/agentregistry_v1.projects.locations.aiApplications.mcpServers.html b/docs/dyn/agentregistry_v1.projects.locations.aiApplications.mcpServers.html new file mode 100644 index 0000000000..9fb7069960 --- /dev/null +++ b/docs/dyn/agentregistry_v1.projects.locations.aiApplications.mcpServers.html @@ -0,0 +1,212 @@ + + + +Agent Registry API . projects . locations . aiApplications . mcpServers
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets details of a single McpServer under an AI Application.
++
+list(parent, pageSize=None, pageToken=None, x__xgafv=None)Lists McpServers under an AI Application.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +get(name, x__xgafv=None)+Gets details of a single McpServer under an AI Application. + +Args: + name: string, Required. Name of the resource. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}/mcpServers/{mcp_server}` (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an MCP (Model Context Protocol) Server. + "attributes": { # Output only. Attributes of the MCP Server. Valid values: * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the MCP Server. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the MCP Server, for example, the GKE Deployment. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. The description of the MCP Server. + "displayName": "A String", # Output only. The display name of the MCP Server. + "interfaces": [ # Output only. The connection details for the MCP Server. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "mcpServerId": "A String", # Output only. A stable, globally unique identifier for MCP Servers. + "name": "A String", # Identifier. The resource name of the MCP Server. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`. + "tools": [ # Output only. Tools provided by the MCP Server. + { # Represents a single tool provided by an MCP Server. + "annotations": { # Annotations describing the characteristics and behavior of a tool or operation. # Output only. Annotations associated with the tool. + "destructiveHint": True or False, # Output only. If true, the tool may perform destructive updates to its environment. If false, the tool performs only additive updates. NOTE: This property is meaningful only when `read_only_hint == false` Default: true + "idempotentHint": True or False, # Output only. If true, calling the tool repeatedly with the same arguments will have no additional effect on its environment. NOTE: This property is meaningful only when `read_only_hint == false` Default: false + "openWorldHint": True or False, # Output only. If true, this tool may interact with an "open world" of external entities. If false, the tool's domain of interaction is closed. For example, the world of a web search tool is open, whereas that of a memory tool is not. Default: true + "readOnlyHint": True or False, # Output only. If true, the tool does not modify its environment. Default: false + "title": "A String", # Output only. A human-readable title for the tool. + }, + "description": "A String", # Output only. Description of what the tool does. + "name": "A String", # Output only. Human-readable name of the tool. + }, + ], + "updateTime": "A String", # Output only. Update time. +}+++ +list(parent, pageSize=None, pageToken=None, x__xgafv=None)+Lists McpServers under an AI Application. + +Args: + parent: string, Required. Parent value (AI Application) for ListAiApplicationMcpServersRequest. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required) + pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default. + pageToken: string, Optional. A token identifying a page of results the server should return. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Message for response to listing McpServers under an AI Application. + "mcpServers": [ # The list of McpServers. + { # Represents an MCP (Model Context Protocol) Server. + "attributes": { # Output only. Attributes of the MCP Server. Valid values: * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the MCP Server. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the MCP Server, for example, the GKE Deployment. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. The description of the MCP Server. + "displayName": "A String", # Output only. The display name of the MCP Server. + "interfaces": [ # Output only. The connection details for the MCP Server. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "mcpServerId": "A String", # Output only. A stable, globally unique identifier for MCP Servers. + "name": "A String", # Identifier. The resource name of the MCP Server. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`. + "tools": [ # Output only. Tools provided by the MCP Server. + { # Represents a single tool provided by an MCP Server. + "annotations": { # Annotations describing the characteristics and behavior of a tool or operation. # Output only. Annotations associated with the tool. + "destructiveHint": True or False, # Output only. If true, the tool may perform destructive updates to its environment. If false, the tool performs only additive updates. NOTE: This property is meaningful only when `read_only_hint == false` Default: true + "idempotentHint": True or False, # Output only. If true, calling the tool repeatedly with the same arguments will have no additional effect on its environment. NOTE: This property is meaningful only when `read_only_hint == false` Default: false + "openWorldHint": True or False, # Output only. If true, this tool may interact with an "open world" of external entities. If false, the tool's domain of interaction is closed. For example, the world of a web search tool is open, whereas that of a memory tool is not. Default: true + "readOnlyHint": True or False, # Output only. If true, the tool does not modify its environment. Default: false + "title": "A String", # Output only. A human-readable title for the tool. + }, + "description": "A String", # Output only. Description of what the tool does. + "name": "A String", # Output only. Human-readable name of the tool. + }, + ], + "updateTime": "A String", # Output only. Update time. + }, + ], + "nextPageToken": "A String", # A token identifying a page of results the server should return. +}+++ + \ No newline at end of file diff --git a/docs/dyn/agentregistry_v1.projects.locations.bindings.html b/docs/dyn/agentregistry_v1.projects.locations.bindings.html index 0e3fe1db1c..389e01e8a8 100644 --- a/docs/dyn/agentregistry_v1.projects.locations.bindings.html +++ b/docs/dyn/agentregistry_v1.projects.locations.bindings.html @@ -128,7 +128,7 @@list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -236,7 +236,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -289,7 +289,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -331,7 +331,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -379,7 +379,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. diff --git a/docs/dyn/agentregistry_v1.projects.locations.services.html b/docs/dyn/agentregistry_v1.projects.locations.services.html index 5818ce7fd7..abc4916585 100644 --- a/docs/dyn/agentregistry_v1.projects.locations.services.html +++ b/docs/dyn/agentregistry_v1.projects.locations.services.html @@ -115,7 +115,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -124,7 +124,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -136,7 +136,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` @@ -229,7 +229,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -238,7 +238,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -250,7 +250,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` @@ -283,7 +283,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -292,7 +292,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -304,7 +304,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` @@ -342,7 +342,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -351,7 +351,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -363,7 +363,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.agents.html b/docs/dyn/agentregistry_v1alpha.projects.locations.agents.html index 16eef2aac2..9720428639 100644 --- a/docs/dyn/agentregistry_v1alpha.projects.locations.agents.html +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.agents.html @@ -259,7 +259,7 @@Method Details
{ # Message for searching Agents "pageSize": 42, # Optional. The maximum number of search results to return per page. The page size is capped at `100`, even if a larger value is specified. A negative value will result in an `INVALID_ARGUMENT` error. If unspecified or set to `0`, a default value of `20` will be used. The server may return fewer results than requested. "pageToken": "A String", # Optional. If present, retrieve the next batch of results from the preceding call to this method. `page_token` must be the value of `next_page_token` from the previous response. The values of all other method parameters, must be identical to those in the previous call. - "searchString": "A String", # Optional. Search criteria used to select the Agents to return. If no search criteria is specified then all accessible Agents will be returned. Search expressions can be used to restrict results based upon searchable fields, where the operators can be used along with the suffix wildcard symbol `*`. See [instructions](https://docs.cloud.google.com/agent-registry/search-agents-and-tools) for more details. Allowed operators: `=`, `:`, `NOT`, `AND`, `OR`, and `()`. Searchable fields: | Field | `=` | `:` | `*` | Keyword Search | |--------------------|-----|-----|-----|----------------| | agentId | Yes | Yes | Yes | Included | | name | No | Yes | Yes | Included | | displayName | No | Yes | Yes | Included | | description | No | Yes | No | Included | | skills | No | Yes | No | Included | | skills.id | No | Yes | No | Included | | skills.name | No | Yes | No | Included | | skills.description | No | Yes | No | Included | | skills.tags | No | Yes | No | Included | | skills.examples | No | Yes | No | Included | Examples: * `agentId="urn:agent:projects-123:projects:123:locations:us-central1:reasoningEngines:1234"` to find the agent with the specified agent ID. * `name:important` to find agents whose name contains `important` as a word. * `displayName:works*` to find agents whose display name contains words that start with `works`. * `skills.tags:test` to find agents whose skills tags contain `test`. * `planner OR booking` to find agents whose metadata contains the words `planner` or `booking`. + "searchString": "A String", # Optional. Search criteria used to select the Agents to return. If no search criteria is specified then all accessible Agents will be returned. Search expressions can be used to restrict results based upon searchable fields, where the operators can be used along with the suffix wildcard symbol `*`. See [instructions](https://docs.cloud.google.com/agent-registry/search-agents-and-tools) for more details. Allowed operators: `=`, `:`, `NOT`, `AND`, `OR`, and `()`. Searchable fields: | Field | `=` | `:` | `*` | Keyword Search | |--------------------|-----|-----|-----|----------------| | agentId | Yes | Yes | Yes | Included | | name | No | Yes | Yes | Included | | displayName | No | Yes | Yes | Included | | description | No | Yes | No | Included | | skills | No | Yes | No | Included | | skills.id | No | Yes | No | Included | | skills.name | No | Yes | No | Included | | skills.description | No | Yes | No | Included | | skills.tags | No | Yes | No | Included | | skills.examples | No | Yes | No | Included | Examples: * `agentId="urn:agent:projects-123:projects:123:locations:us-central1:aiplatform:reasoningEngines:1234"` to find the agent with the specified agent ID. * `name:important` to find agents whose name contains `important` as a word. * `displayName:works*` to find agents whose display name contains words that start with `works`. * `skills.tags:test` to find agents whose skills tags contain `test`. * `planner OR booking` to find agents whose metadata contains the words `planner` or `booking`. } x__xgafv: string, V1 error format. diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.agents.html b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.agents.html new file mode 100644 index 0000000000..360a39d8bb --- /dev/null +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.agents.html @@ -0,0 +1,242 @@ + + + +Agent Registry API . projects . locations . aiApplications . agents
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets details of a single Agent under an AI Application.
++
+list(parent, pageSize=None, pageToken=None, x__xgafv=None)Lists Agents under an AI Application.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +get(name, x__xgafv=None)+Gets details of a single Agent under an AI Application. + +Args: + name: string, Required. Name of the resource. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}/agents/{agent}` (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an Agent. "A2A" below refers to the Agent-to-Agent protocol. + "agentId": "A String", # Output only. A stable, globally unique identifier for agents. + "attributes": { # Output only. Attributes of the Agent. Valid values: * `agentregistry.googleapis.com/system/Framework`: {"framework": "google-adk"} - the agent framework used to develop the Agent. Example values: "google-adk", "langchain", "custom". * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the Agent. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Agent, for example, the Reasoning Engine URI. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "card": { # Full Agent Card payload, often obtained from the A2A Agent Card. # Output only. Full Agent Card payload, when available. + "content": { # Output only. The content of the agent card. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Output only. The type of agent card. + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. The description of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no description. + "displayName": "A String", # Output only. The display name of the agent, often obtained from the A2A Agent Card. + "location": "A String", # Output only. The location where agent is hosted. The value is defined by the hosting environment (i.e. cloud provider). + "name": "A String", # Identifier. The resource name of an Agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. + "protocols": [ # Output only. The connection details for the Agent. + { # Represents the protocol of an Agent. + "interfaces": [ # Output only. The connection details for the Agent. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "protocolVersion": "A String", # Output only. The version of the protocol, for example, the A2A Agent Card version. + "type": "A String", # Output only. The type of the protocol. + }, + ], + "skills": [ # Output only. Skills the agent possesses, often obtained from the A2A Agent Card. + { # Represents the skills of an Agent. + "description": "A String", # Output only. A more detailed description of the skill. + "examples": [ # Output only. Example prompts or scenarios this skill can handle. + "A String", + ], + "id": "A String", # Output only. A unique identifier for the agent's skill. + "name": "A String", # Output only. A human-readable name for the agent's skill. + "tags": [ # Output only. Keywords describing the skill. + "A String", + ], + }, + ], + "uid": "A String", # Output only. A universally unique identifier for the Agent. + "updateTime": "A String", # Output only. Update time. + "version": "A String", # Output only. The version of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no version or agent is not an A2A Agent. +}+++ +list(parent, pageSize=None, pageToken=None, x__xgafv=None)+Lists Agents under an AI Application. + +Args: + parent: string, Required. Parent value (AI Application) for ListAiApplicationAgentsRequest. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required) + pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default. + pageToken: string, Optional. A token identifying a page of results the server should return. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Message for response to listing Agents under an AI Application. + "agents": [ # The list of Agents. + { # Represents an Agent. "A2A" below refers to the Agent-to-Agent protocol. + "agentId": "A String", # Output only. A stable, globally unique identifier for agents. + "attributes": { # Output only. Attributes of the Agent. Valid values: * `agentregistry.googleapis.com/system/Framework`: {"framework": "google-adk"} - the agent framework used to develop the Agent. Example values: "google-adk", "langchain", "custom". * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the Agent. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Agent, for example, the Reasoning Engine URI. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "card": { # Full Agent Card payload, often obtained from the A2A Agent Card. # Output only. Full Agent Card payload, when available. + "content": { # Output only. The content of the agent card. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Output only. The type of agent card. + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. The description of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no description. + "displayName": "A String", # Output only. The display name of the agent, often obtained from the A2A Agent Card. + "location": "A String", # Output only. The location where agent is hosted. The value is defined by the hosting environment (i.e. cloud provider). + "name": "A String", # Identifier. The resource name of an Agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. + "protocols": [ # Output only. The connection details for the Agent. + { # Represents the protocol of an Agent. + "interfaces": [ # Output only. The connection details for the Agent. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "protocolVersion": "A String", # Output only. The version of the protocol, for example, the A2A Agent Card version. + "type": "A String", # Output only. The type of the protocol. + }, + ], + "skills": [ # Output only. Skills the agent possesses, often obtained from the A2A Agent Card. + { # Represents the skills of an Agent. + "description": "A String", # Output only. A more detailed description of the skill. + "examples": [ # Output only. Example prompts or scenarios this skill can handle. + "A String", + ], + "id": "A String", # Output only. A unique identifier for the agent's skill. + "name": "A String", # Output only. A human-readable name for the agent's skill. + "tags": [ # Output only. Keywords describing the skill. + "A String", + ], + }, + ], + "uid": "A String", # Output only. A universally unique identifier for the Agent. + "updateTime": "A String", # Output only. Update time. + "version": "A String", # Output only. The version of the Agent, often obtained from the A2A Agent Card. Empty if Agent Card has no version or agent is not an A2A Agent. + }, + ], + "nextPageToken": "A String", # A token identifying a page of results the server should return. +}+++ + \ No newline at end of file diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.endpoints.html b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.endpoints.html new file mode 100644 index 0000000000..600eb1bd57 --- /dev/null +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.endpoints.html @@ -0,0 +1,186 @@ + + + +list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Agent Registry API . projects . locations . aiApplications . endpoints
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets details of a single Endpoint under an AI Application.
++
+list(parent, pageSize=None, pageToken=None, x__xgafv=None)Lists Endpoints under an AI Application.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +get(name, x__xgafv=None)+Gets details of a single Endpoint under an AI Application. + +Args: + name: string, Required. Name of the resource. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}/endpoints/{endpoint}` (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an Endpoint. + "attributes": { # Output only. Attributes of the Endpoint. Valid values: * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Endpoint, for example, the GKE Deployment. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. Description of an Endpoint. + "displayName": "A String", # Output only. Display name for the Endpoint. + "endpointId": "A String", # Output only. A stable, globally unique identifier for Endpoint. + "interfaces": [ # Required. The connection details for the Endpoint. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "name": "A String", # Identifier. The resource name of the Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`. + "updateTime": "A String", # Output only. Update time. +}+++ +list(parent, pageSize=None, pageToken=None, x__xgafv=None)+Lists Endpoints under an AI Application. + +Args: + parent: string, Required. Parent value (AI Application) for ListAiApplicationEndpointsRequest. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required) + pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default. + pageToken: string, Optional. A token identifying a page of results the server should return. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Message for response to listing Endpoints under an AI Application. + "endpoints": [ # The list of Endpoints. + { # Represents an Endpoint. + "attributes": { # Output only. Attributes of the Endpoint. Valid values: * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the Endpoint, for example, the GKE Deployment. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. Description of an Endpoint. + "displayName": "A String", # Output only. Display name for the Endpoint. + "endpointId": "A String", # Output only. A stable, globally unique identifier for Endpoint. + "interfaces": [ # Required. The connection details for the Endpoint. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "name": "A String", # Identifier. The resource name of the Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`. + "updateTime": "A String", # Output only. Update time. + }, + ], + "nextPageToken": "A String", # A token identifying a page of results the server should return. +}+++ + \ No newline at end of file diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.html b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.html index 10caa9bff9..a662a23753 100644 --- a/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.html +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.html @@ -74,12 +74,36 @@list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Agent Registry API . projects . locations . aiApplications
Instance Methods
++
+agents()+Returns the agents Resource.
+ ++
+endpoints()+Returns the endpoints Resource.
+ ++
+mcpServers()+Returns the mcpServers Resource.
+Close httplib2 connections.
+ +Gets details of a single AI Application.
getIamPolicy(resource, options_requestedPolicyVersion=None, x__xgafv=None)Gets the access control policy for a resource. Returns an empty policy if the resource exists and does not have a policy set.
++
+list(parent, filter=None, orderBy=None, pageSize=None, pageToken=None, x__xgafv=None)Lists AI Applications in a given project and location.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
setIamPolicy(resource, body=None, x__xgafv=None)Sets the access control policy on the specified resource. Replaces any existing policy. Can return `NOT_FOUND`, `INVALID_ARGUMENT`, and `PERMISSION_DENIED` errors.
@@ -92,6 +116,78 @@Method Details
Close httplib2 connections.
get(name, x__xgafv=None)
+ Gets details of a single AI Application.
+
+Args:
+ name: string, Required. Target AI Application resource name. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Represents an AI Application.
+ "applicationProperties": { # Additional system properties of an Application. # Output only. Properties of an underlying cloud resource that can comprise an AI Application.
+ "extendedMetadata": { # Output only. Additional metadata specific to the App Hub application. The key is a string that identifies the type of metadata and the value is the metadata contents specific to that type. Key format: `apphub.googleapis.com/{metadataType}`
+ "a_key": { # Additional metadata for a Service or Workload.
+ "metadataStruct": { # Output only. The metadata contents.
+ "a_key": "", # Properties of the object.
+ },
+ },
+ },
+ },
+ "attributes": { # Consumer provided attributes. # Optional. Consumer provided attributes.
+ "businessOwners": [ # Optional. Business team that ensures user needs are met and value is delivered
+ { # Contact information of stakeholders.
+ "channel": { # Separate message to accommodate custom formats across IRC and Slack. # Optional. Communication channel of the contacts.
+ "uri": "A String", # Required. URI of the channel.
+ },
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "criticality": { # Criticality of the Application, Service, or Workload # Optional. User-defined criticality information.
+ "level": "A String", # Optional. Criticality level. Can contain only lowercase letters, numeric characters, underscores, and dashes. Can have a maximum length of 63 characters. Deprecated: Please refer to type instead.
+ "missionCritical": True or False, # Optional. Indicates mission-critical Application, Service, or Workload. Deprecated: Please refer to type instead.
+ "type": "A String", # Required. Criticality Type.
+ },
+ "developerOwners": [ # Optional. Developer team that owns development and coding.
+ { # Contact information of stakeholders.
+ "channel": { # Separate message to accommodate custom formats across IRC and Slack. # Optional. Communication channel of the contacts.
+ "uri": "A String", # Required. URI of the channel.
+ },
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "environment": { # Environment of the Application, Service, or Workload # Optional. User-defined environment information.
+ "environment": "A String", # Optional. Environment name. Can contain only lowercase letters, numeric characters, underscores, and dashes. Can have a maximum length of 63 characters. Deprecated: Please refer to type instead.
+ "type": "A String", # Required. Environment Type.
+ },
+ "operatorOwners": [ # Optional. Operator team that ensures runtime and operations.
+ { # Contact information of stakeholders.
+ "channel": { # Separate message to accommodate custom formats across IRC and Slack. # Optional. Communication channel of the contacts.
+ "uri": "A String", # Required. URI of the channel.
+ },
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ },
+ "createTime": "A String", # Output only. Creation time.
+ "description": "A String", # Optional. User-defined description of the AI Application.
+ "displayName": "A String", # Optional. User-defined name for the AI Application.
+ "name": "A String", # Identifier. Resource name of the AI Application. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}`
+ "state": "A String", # Output only. The current state of the AI Application.
+ "uid": "A String", # Output only. Universally unique identifier (UUID4) for the AI Application.
+ "updateTime": "A String", # Output only. Last update time.
+}
+getIamPolicy(resource, options_requestedPolicyVersion=None, x__xgafv=None)
Gets the access control policy for a resource. Returns an empty policy if the resource exists and does not have a policy set. @@ -140,6 +236,104 @@Method Details
}
list(parent, filter=None, orderBy=None, pageSize=None, pageToken=None, x__xgafv=None)
+ Lists AI Applications in a given project and location.
+
+Args:
+ parent: string, Required. Parent project and location to query. Format: `projects/{project}/locations/{location}` (required)
+ filter: string, Optional. Filtering results.
+ orderBy: string, Optional. Hint for how to order the results.
+ pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default.
+ pageToken: string, Optional. A token identifying a page of results the server should return.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Message for response to listing AiApplications.
+ "aiApplications": [ # The list of AiApplications.
+ { # Represents an AI Application.
+ "applicationProperties": { # Additional system properties of an Application. # Output only. Properties of an underlying cloud resource that can comprise an AI Application.
+ "extendedMetadata": { # Output only. Additional metadata specific to the App Hub application. The key is a string that identifies the type of metadata and the value is the metadata contents specific to that type. Key format: `apphub.googleapis.com/{metadataType}`
+ "a_key": { # Additional metadata for a Service or Workload.
+ "metadataStruct": { # Output only. The metadata contents.
+ "a_key": "", # Properties of the object.
+ },
+ },
+ },
+ },
+ "attributes": { # Consumer provided attributes. # Optional. Consumer provided attributes.
+ "businessOwners": [ # Optional. Business team that ensures user needs are met and value is delivered
+ { # Contact information of stakeholders.
+ "channel": { # Separate message to accommodate custom formats across IRC and Slack. # Optional. Communication channel of the contacts.
+ "uri": "A String", # Required. URI of the channel.
+ },
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "criticality": { # Criticality of the Application, Service, or Workload # Optional. User-defined criticality information.
+ "level": "A String", # Optional. Criticality level. Can contain only lowercase letters, numeric characters, underscores, and dashes. Can have a maximum length of 63 characters. Deprecated: Please refer to type instead.
+ "missionCritical": True or False, # Optional. Indicates mission-critical Application, Service, or Workload. Deprecated: Please refer to type instead.
+ "type": "A String", # Required. Criticality Type.
+ },
+ "developerOwners": [ # Optional. Developer team that owns development and coding.
+ { # Contact information of stakeholders.
+ "channel": { # Separate message to accommodate custom formats across IRC and Slack. # Optional. Communication channel of the contacts.
+ "uri": "A String", # Required. URI of the channel.
+ },
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ "environment": { # Environment of the Application, Service, or Workload # Optional. User-defined environment information.
+ "environment": "A String", # Optional. Environment name. Can contain only lowercase letters, numeric characters, underscores, and dashes. Can have a maximum length of 63 characters. Deprecated: Please refer to type instead.
+ "type": "A String", # Required. Environment Type.
+ },
+ "operatorOwners": [ # Optional. Operator team that ensures runtime and operations.
+ { # Contact information of stakeholders.
+ "channel": { # Separate message to accommodate custom formats across IRC and Slack. # Optional. Communication channel of the contacts.
+ "uri": "A String", # Required. URI of the channel.
+ },
+ "displayName": "A String", # Optional. Contact's name. Can have a maximum length of 63 characters.
+ "email": "A String", # Required. Email address of the contacts.
+ },
+ ],
+ },
+ "createTime": "A String", # Output only. Creation time.
+ "description": "A String", # Optional. User-defined description of the AI Application.
+ "displayName": "A String", # Optional. User-defined name for the AI Application.
+ "name": "A String", # Identifier. Resource name of the AI Application. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}`
+ "state": "A String", # Output only. The current state of the AI Application.
+ "uid": "A String", # Output only. Universally unique identifier (UUID4) for the AI Application.
+ "updateTime": "A String", # Output only. Last update time.
+ },
+ ],
+ "nextPageToken": "A String", # A token identifying a page of results the server should return.
+ "unreachable": [ # Unordered list. Locations that could not be reached.
+ "A String",
+ ],
+}
+list_next(previous_request, previous_response)
+ Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++
setIamPolicy(resource, body=None, x__xgafv=None)
Sets the access control policy on the specified resource. Replaces any existing policy. Can return `NOT_FOUND`, `INVALID_ARGUMENT`, and `PERMISSION_DENIED` errors. diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.mcpServers.html b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.mcpServers.html new file mode 100644 index 0000000000..5031043237 --- /dev/null +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.aiApplications.mcpServers.html @@ -0,0 +1,212 @@ + + + +-Agent Registry API . projects . locations . aiApplications . mcpServers
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets details of a single McpServer under an AI Application.
++
+list(parent, pageSize=None, pageToken=None, x__xgafv=None)Lists McpServers under an AI Application.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +get(name, x__xgafv=None)+Gets details of a single McpServer under an AI Application. + +Args: + name: string, Required. Name of the resource. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}/mcpServers/{mcp_server}` (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an MCP (Model Context Protocol) Server. + "attributes": { # Output only. Attributes of the MCP Server. Valid values: * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the MCP Server. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the MCP Server, for example, the GKE Deployment. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. The description of the MCP Server. + "displayName": "A String", # Output only. The display name of the MCP Server. + "interfaces": [ # Output only. The connection details for the MCP Server. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "mcpServerId": "A String", # Output only. A stable, globally unique identifier for MCP Servers. + "name": "A String", # Identifier. The resource name of the MCP Server. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`. + "tools": [ # Output only. Tools provided by the MCP Server. + { # Represents a single tool provided by an MCP Server. + "annotations": { # Annotations describing the characteristics and behavior of a tool or operation. # Output only. Annotations associated with the tool. + "destructiveHint": True or False, # Output only. If true, the tool may perform destructive updates to its environment. If false, the tool performs only additive updates. NOTE: This property is meaningful only when `read_only_hint == false` Default: true + "idempotentHint": True or False, # Output only. If true, calling the tool repeatedly with the same arguments will have no additional effect on its environment. NOTE: This property is meaningful only when `read_only_hint == false` Default: false + "openWorldHint": True or False, # Output only. If true, this tool may interact with an "open world" of external entities. If false, the tool's domain of interaction is closed. For example, the world of a web search tool is open, whereas that of a memory tool is not. Default: true + "readOnlyHint": True or False, # Output only. If true, the tool does not modify its environment. Default: false + "title": "A String", # Output only. A human-readable title for the tool. + }, + "description": "A String", # Output only. Description of what the tool does. + "name": "A String", # Output only. Human-readable name of the tool. + }, + ], + "updateTime": "A String", # Output only. Update time. +}+++ +list(parent, pageSize=None, pageToken=None, x__xgafv=None)+Lists McpServers under an AI Application. + +Args: + parent: string, Required. Parent value (AI Application) for ListAiApplicationMcpServersRequest. Format: `projects/{project}/locations/{location}/aiApplications/{ai_application}` (required) + pageSize: integer, Optional. Requested page size. Server may return fewer items than requested. If unspecified, server will pick an appropriate default. + pageToken: string, Optional. A token identifying a page of results the server should return. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Message for response to listing McpServers under an AI Application. + "mcpServers": [ # The list of McpServers. + { # Represents an MCP (Model Context Protocol) Server. + "attributes": { # Output only. Attributes of the MCP Server. Valid values: * `agentregistry.googleapis.com/system/RuntimeIdentity`: {"principal": "principal://..."} - the runtime identity associated with the MCP Server. * `agentregistry.googleapis.com/system/RuntimeReference`: {"uri": "//..."} - the URI of the underlying resource hosting the MCP Server, for example, the GKE Deployment. + "a_key": { + "a_key": "", # Properties of the object. + }, + }, + "createTime": "A String", # Output only. Create time. + "description": "A String", # Output only. The description of the MCP Server. + "displayName": "A String", # Output only. The display name of the MCP Server. + "interfaces": [ # Output only. The connection details for the MCP Server. + { # Represents the connection details for an Agent or MCP Server. + "protocolBinding": "A String", # Required. The protocol binding of the interface. + "url": "A String", # Required. The destination URL. + }, + ], + "mcpServerId": "A String", # Output only. A stable, globally unique identifier for MCP Servers. + "name": "A String", # Identifier. The resource name of the MCP Server. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`. + "tools": [ # Output only. Tools provided by the MCP Server. + { # Represents a single tool provided by an MCP Server. + "annotations": { # Annotations describing the characteristics and behavior of a tool or operation. # Output only. Annotations associated with the tool. + "destructiveHint": True or False, # Output only. If true, the tool may perform destructive updates to its environment. If false, the tool performs only additive updates. NOTE: This property is meaningful only when `read_only_hint == false` Default: true + "idempotentHint": True or False, # Output only. If true, calling the tool repeatedly with the same arguments will have no additional effect on its environment. NOTE: This property is meaningful only when `read_only_hint == false` Default: false + "openWorldHint": True or False, # Output only. If true, this tool may interact with an "open world" of external entities. If false, the tool's domain of interaction is closed. For example, the world of a web search tool is open, whereas that of a memory tool is not. Default: true + "readOnlyHint": True or False, # Output only. If true, the tool does not modify its environment. Default: false + "title": "A String", # Output only. A human-readable title for the tool. + }, + "description": "A String", # Output only. Description of what the tool does. + "name": "A String", # Output only. Human-readable name of the tool. + }, + ], + "updateTime": "A String", # Output only. Update time. + }, + ], + "nextPageToken": "A String", # A token identifying a page of results the server should return. +}+++ + \ No newline at end of file diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.bindings.html b/docs/dyn/agentregistry_v1alpha.projects.locations.bindings.html index ff42f302ba..2026487bb3 100644 --- a/docs/dyn/agentregistry_v1alpha.projects.locations.bindings.html +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.bindings.html @@ -128,7 +128,7 @@list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -236,7 +236,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -289,7 +289,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -331,7 +331,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. @@ -379,7 +379,7 @@Method Details
"description": "A String", # Optional. User-defined description of a Binding. Can have a maximum length of `2048` characters. "displayName": "A String", # Optional. User-defined display name for the Binding. Can have a maximum length of `63` characters. "name": "A String", # Required. Identifier. The resource name of the Binding. Format: `projects/{project}/locations/{location}/bindings/{binding}`. - "source": { # The source of the Binding. # Required. The target Agent of the Binding. + "source": { # The source of the Binding. # Optional. The source Agent of the Binding. "identifier": "A String", # The identifier of the source Agent. Format: * `urn:agent:{publisher}:{namespace}:{name}` }, "target": { # The target of the Binding. # Required. The target Agent Registry Resource of the Binding. diff --git a/docs/dyn/agentregistry_v1alpha.projects.locations.services.html b/docs/dyn/agentregistry_v1alpha.projects.locations.services.html index 52053953d5..def28fa7b6 100644 --- a/docs/dyn/agentregistry_v1alpha.projects.locations.services.html +++ b/docs/dyn/agentregistry_v1alpha.projects.locations.services.html @@ -115,7 +115,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -124,7 +124,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -136,7 +136,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` @@ -229,7 +229,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -238,7 +238,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -250,7 +250,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` @@ -283,7 +283,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -292,7 +292,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -304,7 +304,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` @@ -342,7 +342,7 @@Method Details
"content": { # Optional. The content of the Agent spec in the JSON format. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the agent spec content. + "type": "A String", # Required. Immutable. The type of the agent spec content. }, "createTime": "A String", # Output only. Create time. "description": "A String", # Optional. User-defined description of an Service. Can have a maximum length of `2048` characters. @@ -351,7 +351,7 @@Method Details
"content": { # Optional. The content of the endpoint spec. Reserved for future use. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the endpoint spec content. + "type": "A String", # Required. Immutable. The type of the endpoint spec content. }, "interfaces": [ # Optional. The connection details for the Service. { # Represents the connection details for an Agent or MCP Server. @@ -363,7 +363,7 @@Method Details
"content": { # Optional. The content of the MCP Server spec. This payload is validated against the schema for the specified type. The content size is limited to `10KB`. "a_key": "", # Properties of the object. }, - "type": "A String", # Required. The type of the MCP Server spec content. + "type": "A String", # Required. Immutable. The type of the MCP Server spec content. }, "name": "A String", # Identifier. The resource name of the Service. Format: `projects/{project}/locations/{location}/services/{service}`. "registryResource": "A String", # Output only. The resource name of the resulting Agent, MCP Server, or Endpoint. Format: * `projects/{project}/locations/{location}/mcpServers/{mcp_server}` * `projects/{project}/locations/{location}/agents/{agent}` * `projects/{project}/locations/{location}/endpoints/{endpoint}` diff --git a/docs/dyn/aiplatform_v1.endpoints.html b/docs/dyn/aiplatform_v1.endpoints.html index 31ff268ca6..dd523e512c 100644 --- a/docs/dyn/aiplatform_v1.endpoints.html +++ b/docs/dyn/aiplatform_v1.endpoints.html @@ -621,13 +621,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1316,13 +1316,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2349,13 +2349,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.agents.html b/docs/dyn/aiplatform_v1.projects.locations.agents.html index 7552ad1069..9b4c5b8964 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.agents.html +++ b/docs/dyn/aiplatform_v1.projects.locations.agents.html @@ -128,6 +128,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. @@ -236,6 +240,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. @@ -321,6 +329,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. @@ -377,6 +389,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. diff --git a/docs/dyn/aiplatform_v1.projects.locations.cachedContents.html b/docs/dyn/aiplatform_v1.projects.locations.cachedContents.html index 505564a4fe..86426d669e 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.cachedContents.html +++ b/docs/dyn/aiplatform_v1.projects.locations.cachedContents.html @@ -327,13 +327,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -794,13 +794,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1286,13 +1286,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1764,13 +1764,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2250,13 +2250,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2718,13 +2718,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.endpoints.html b/docs/dyn/aiplatform_v1.projects.locations.endpoints.html index 5f30291d96..5b7d0bf27b 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.endpoints.html +++ b/docs/dyn/aiplatform_v1.projects.locations.endpoints.html @@ -694,13 +694,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2285,13 +2285,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4800,13 +4800,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.evaluationItems.html b/docs/dyn/aiplatform_v1.projects.locations.evaluationItems.html index be9727f283..dd6d3e07f7 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.evaluationItems.html +++ b/docs/dyn/aiplatform_v1.projects.locations.evaluationItems.html @@ -120,7 +120,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -137,13 +137,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -378,13 +378,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -743,13 +743,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -984,13 +984,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1348,13 +1348,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1589,13 +1589,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2087,7 +2087,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -2104,13 +2104,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2345,13 +2345,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2710,13 +2710,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2951,13 +2951,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3315,13 +3315,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3556,13 +3556,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4045,7 +4045,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -4062,13 +4062,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4303,13 +4303,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4668,13 +4668,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4909,13 +4909,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5273,13 +5273,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5514,13 +5514,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6012,7 +6012,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -6029,13 +6029,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6270,13 +6270,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6635,13 +6635,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6876,13 +6876,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7240,13 +7240,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7481,13 +7481,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8012,7 +8012,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -8029,13 +8029,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8270,13 +8270,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8635,13 +8635,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8876,13 +8876,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9240,13 +9240,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9481,13 +9481,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9979,7 +9979,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -9996,13 +9996,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -10237,13 +10237,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -10602,13 +10602,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -10843,13 +10843,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11207,13 +11207,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11448,13 +11448,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11950,7 +11950,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -11967,13 +11967,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12208,13 +12208,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12573,13 +12573,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12814,13 +12814,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13178,13 +13178,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13419,13 +13419,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13917,7 +13917,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -13934,13 +13934,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14175,13 +14175,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14540,13 +14540,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14781,13 +14781,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -15145,13 +15145,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -15386,13 +15386,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.evaluationRuns.html b/docs/dyn/aiplatform_v1.projects.locations.evaluationRuns.html index 60681c90e3..7fec7186b9 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.evaluationRuns.html +++ b/docs/dyn/aiplatform_v1.projects.locations.evaluationRuns.html @@ -167,6 +167,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -1955,13 +1956,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2412,6 +2413,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -4200,13 +4202,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4699,6 +4701,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -6487,13 +6490,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6957,6 +6960,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -8745,13 +8749,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.evaluationSets.html b/docs/dyn/aiplatform_v1.projects.locations.evaluationSets.html index 69cb4acfc5..05167b91c3 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.evaluationSets.html +++ b/docs/dyn/aiplatform_v1.projects.locations.evaluationSets.html @@ -127,13 +127,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -395,13 +395,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -705,13 +705,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1007,13 +1007,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1342,13 +1342,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1629,13 +1629,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1898,13 +1898,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.html b/docs/dyn/aiplatform_v1.projects.locations.html index 15ed854b31..4816b92152 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.html +++ b/docs/dyn/aiplatform_v1.projects.locations.html @@ -398,13 +398,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -696,13 +696,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2634,13 +2634,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2981,13 +2981,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3418,13 +3418,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3659,13 +3659,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6209,13 +6209,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6809,7 +6809,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -6826,13 +6826,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7067,13 +7067,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7432,13 +7432,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7673,13 +7673,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8037,13 +8037,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8278,13 +8278,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9033,13 +9033,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.publishers.models.html b/docs/dyn/aiplatform_v1.projects.locations.publishers.models.html index af9ee3c61c..1ec1a77554 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.publishers.models.html +++ b/docs/dyn/aiplatform_v1.projects.locations.publishers.models.html @@ -628,13 +628,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1491,13 +1491,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2752,13 +2752,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.projects.locations.semanticGovernancePolicies.html b/docs/dyn/aiplatform_v1.projects.locations.semanticGovernancePolicies.html index a579b76b3f..eaf45b27e0 100644 --- a/docs/dyn/aiplatform_v1.projects.locations.semanticGovernancePolicies.html +++ b/docs/dyn/aiplatform_v1.projects.locations.semanticGovernancePolicies.html @@ -116,7 +116,7 @@Method Details
The object takes the form of: { # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent. - "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. + "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}` "agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID` "agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request. "denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters. @@ -129,7 +129,7 @@Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy. { # Represents a governance policy applied to MCP tools used by an Agent. "mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}` - "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer. + "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified. "A String", ], }, @@ -220,7 +220,7 @@Method Details
An object of the form: { # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent. - "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. + "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}` "agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID` "agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request. "denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters. @@ -233,7 +233,7 @@Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy. { # Represents a governance policy applied to MCP tools used by an Agent. "mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}` - "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer. + "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified. "A String", ], }, @@ -264,7 +264,7 @@Method Details
"nextPageToken": "A String", # A token to retrieve the next page of results. Pass to ListSemanticGovernancePoliciesRequest.page_token to obtain that page. "semanticGovernancePolicies": [ # The list of SemanticGovernancePolicies. { # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent. - "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. + "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}` "agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID` "agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request. "denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters. @@ -277,7 +277,7 @@Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy. { # Represents a governance policy applied to MCP tools used by an Agent. "mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}` - "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer. + "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified. "A String", ], }, @@ -314,7 +314,7 @@Method Details
The object takes the form of: { # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent. - "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. + "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}` "agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID` "agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request. "denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters. @@ -327,7 +327,7 @@Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy. { # Represents a governance policy applied to MCP tools used by an Agent. "mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}` - "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer. + "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.publishers.models.html b/docs/dyn/aiplatform_v1.publishers.models.html index df3743f314..079feb45b2 100644 --- a/docs/dyn/aiplatform_v1.publishers.models.html +++ b/docs/dyn/aiplatform_v1.publishers.models.html @@ -614,13 +614,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1309,13 +1309,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2917,13 +2917,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1.v1.html b/docs/dyn/aiplatform_v1.v1.html index 54a2338226..9f4b083d7a 100644 --- a/docs/dyn/aiplatform_v1.v1.html +++ b/docs/dyn/aiplatform_v1.v1.html @@ -1112,13 +1112,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1459,13 +1459,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1896,13 +1896,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2137,13 +2137,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4686,13 +4686,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.endpoints.html b/docs/dyn/aiplatform_v1beta1.endpoints.html index 9ee4128221..2c2a6c69ab 100644 --- a/docs/dyn/aiplatform_v1beta1.endpoints.html +++ b/docs/dyn/aiplatform_v1beta1.endpoints.html @@ -627,13 +627,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1334,13 +1334,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2380,13 +2380,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.agentAnomalyDetectionScopes.html b/docs/dyn/aiplatform_v1beta1.projects.locations.agentAnomalyDetectionScopes.html index b907c47fb2..fa84cfd5cd 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.agentAnomalyDetectionScopes.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.agentAnomalyDetectionScopes.html @@ -108,6 +108,7 @@Method Details
The object takes the form of: { # Represents an Agent Anomaly Detection Scope. + "autoEnroll": True or False, # Optional. When true, agents created in Agent Engine that share log and observability buckets with this scope are automatically enrolled as MonitoredAgents. When false, agents must be enrolled explicitly via MonitoredAgent operations. "displayName": "A String", # Optional. User-provided display name of the AgentAnomalyDetectionScope. "logBuckets": [ # Required. Customer-owned Cloud Logging bucket resource names attached to this scope. Format: `projects/{project}/locations/{location}/buckets/{bucket}`. "A String", @@ -199,6 +200,7 @@Method Details
An object of the form: { # Represents an Agent Anomaly Detection Scope. + "autoEnroll": True or False, # Optional. When true, agents created in Agent Engine that share log and observability buckets with this scope are automatically enrolled as MonitoredAgents. When false, agents must be enrolled explicitly via MonitoredAgent operations. "displayName": "A String", # Optional. User-provided display name of the AgentAnomalyDetectionScope. "logBuckets": [ # Required. Customer-owned Cloud Logging bucket resource names attached to this scope. Format: `projects/{project}/locations/{location}/buckets/{bucket}`. "A String", @@ -230,6 +232,7 @@Method Details
{ # Response message for AgentAnomalyDetectionService.ListAgentAnomalyDetectionScopes. "agentAnomalyDetectionScopes": [ # A list of AgentAnomalyDetectionScopes that matches the specified filter in the request. { # Represents an Agent Anomaly Detection Scope. + "autoEnroll": True or False, # Optional. When true, agents created in Agent Engine that share log and observability buckets with this scope are automatically enrolled as MonitoredAgents. When false, agents must be enrolled explicitly via MonitoredAgent operations. "displayName": "A String", # Optional. User-provided display name of the AgentAnomalyDetectionScope. "logBuckets": [ # Required. Customer-owned Cloud Logging bucket resource names attached to this scope. Format: `projects/{project}/locations/{location}/buckets/{bucket}`. "A String", diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.agents.html b/docs/dyn/aiplatform_v1beta1.projects.locations.agents.html index 6c717d0896..bb09bc3823 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.agents.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.agents.html @@ -133,6 +133,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. @@ -241,6 +245,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. @@ -326,6 +334,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. @@ -382,6 +394,10 @@Method Details
"metadata": { # Optional. The metadata for the agent. "a_key": "A String", }, + "modelConfig": { # Selects the model configuration for an agent. Starts with just the model id/alias; extensible with further model parameters without a breaking change. # Optional. Model configuration for the agent. When set, tasks running under this agent default to this model configuration if they do not specify their own. + "model": "A String", # Optional. Publisher model resource name (an `aiplatform.googleapis.com/PublisherModel`) of the form "publishers/{publisher}/models/{model}", e.g. "publishers/anthropic/models/claude-opus-4-8" or "publishers/google/models/gemini-3-flash". Empty means the agent's default gemini model. + "temperature": 3.14, # Optional. Sampling temperature for the model's generation config (valid range (0.0, 2.0]); a value outside that range leaves the backend default standing. + }, "name": "A String", # Identifier. The resource name of the agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. "object": "A String", # Output only. The object type of the resource. For agents, the value is `agent`. "observabilityConfig": { # Observability settings for one agent's sessions. # Optional. Observability settings for this agent's sessions. diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.cachedContents.html b/docs/dyn/aiplatform_v1beta1.projects.locations.cachedContents.html index d1e484097e..24c86e2a06 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.cachedContents.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.cachedContents.html @@ -329,13 +329,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -805,13 +805,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1306,13 +1306,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1793,13 +1793,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2288,13 +2288,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2765,13 +2765,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.datasets.html b/docs/dyn/aiplatform_v1beta1.projects.locations.datasets.html index d1c56037e3..3b8515aed3 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.datasets.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.datasets.html @@ -561,13 +561,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1252,13 +1252,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.endpoints.html b/docs/dyn/aiplatform_v1beta1.projects.locations.endpoints.html index 5718dbbbff..b18932bdba 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.endpoints.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.endpoints.html @@ -709,13 +709,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2526,13 +2526,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5397,13 +5397,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationItems.html b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationItems.html index 95fe117bf8..e807a2b018 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationItems.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationItems.html @@ -125,7 +125,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -142,13 +142,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -390,13 +390,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -859,13 +859,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1107,13 +1107,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1575,13 +1575,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1823,13 +1823,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2330,7 +2330,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -2347,13 +2347,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2595,13 +2595,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3064,13 +3064,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3312,13 +3312,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3780,13 +3780,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4028,13 +4028,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4526,7 +4526,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -4543,13 +4543,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4791,13 +4791,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5260,13 +5260,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5508,13 +5508,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5976,13 +5976,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6224,13 +6224,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6731,7 +6731,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -6748,13 +6748,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6996,13 +6996,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7465,13 +7465,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7713,13 +7713,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8181,13 +8181,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8429,13 +8429,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8969,7 +8969,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -8986,13 +8986,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9234,13 +9234,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9703,13 +9703,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9951,13 +9951,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -10419,13 +10419,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -10667,13 +10667,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11174,7 +11174,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -11191,13 +11191,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11439,13 +11439,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11908,13 +11908,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12156,13 +12156,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12624,13 +12624,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12872,13 +12872,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13383,7 +13383,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationItemType": "A String", # Required. The type of the EvaluationItem. - "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # The request to evaluate. + "evaluationRequest": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # The request to evaluate. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -13400,13 +13400,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13648,13 +13648,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14117,13 +14117,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14365,13 +14365,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14833,13 +14833,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -15081,13 +15081,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -15588,7 +15588,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -15605,13 +15605,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -15853,13 +15853,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -16322,13 +16322,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -16570,13 +16570,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -17038,13 +17038,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -17286,13 +17286,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationMetrics.html b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationMetrics.html index 559286f9e4..bc2bc9d6f2 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationMetrics.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationMetrics.html @@ -327,6 +327,183 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. @@ -552,83 +729,260 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, - }, - "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. - "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. - "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. - }, - }, - "name": "A String", # Identifier. The resource name of the EvaluationMetric. Format: `projects/{project}/locations/{location}/evaluationMetrics/{evaluation_metric}` - "updateTime": "A String", # Output only. The time when the EvaluationMetric was last updated. -} - - evaluationMetricId: string, Optional. The ID to use for the EvaluationMetric, which will become the final component of the EvaluationMetric's resource name. This value should be 1-63 characters, and valid characters are /a-z-/. The first character must be a lowercase letter, and the last character must be a lowercase letter or number. - x__xgafv: string, V1 error format. - Allowed values - 1 - v1 error format - 2 - v2 error format - -Returns: - An object of the form: - - { # EvaluationMetric is a resource that represents a reusable metric configuration. - "createTime": "A String", # Output only. The time when the EvaluationMetric was created. - "description": "A String", # Optional. A description of the EvaluationMetric. - "displayName": "A String", # Required. The user-friendly display name for the EvaluationMetric. - "encryptionSpec": { # Represents a customer-managed encryption key specification that can be applied to a Vertex AI resource. # Optional. Customer-managed encryption key spec for this EvaluationMetric. If set, this EvaluationMetric will be secured by this key. - "kmsKeyName": "A String", # Required. Resource name of the Cloud KMS key used to protect the resource. The Cloud KMS key must be in the same region as the resource. It must have the format `projects/{project}/locations/{location}/keyRings/{key_ring}/cryptoKeys/{crypto_key}`. - }, - "gcsUri": "A String", # Optional. The Google Cloud Storage URI that stores the metric specification.. - "labels": { # Optional. Labels for the evaluation metric. - "a_key": "A String", - }, - "metric": { # The metric used for running evaluations. # Optional. The metric configuration. Only LLMMetric and CustomCodeExecutionMetric are supported. - "aggregationMetrics": [ # Optional. The aggregation metrics to use. - "A String", - ], - "bleuSpec": { # Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. # Spec for bleu metric. - "useEffectiveOrder": True or False, # Optional. Whether to use_effective_order to compute bleu score. - }, - "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. - "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - "a_key": "", # Properties of the object. - }, - "type": "A String", # Required. The type of the computation based metric. - }, - "customCodeExecutionSpec": { # Specificies a metric that is populated by evaluating user-defined Python code. # Spec for Custom Code Execution metric. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used; requests from regions where the sandbox is unavailable will fail with UNIMPLEMENTED. - "evaluationFunction": "A String", # Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. - }, - "exactMatchSpec": { # Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. # Spec for exact match metric. - }, - "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. - "additionalConfig": { # Optional. Optional additional configuration for the metric. - "a_key": "", # Properties of the object. - }, - "judgeAutoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Optional configuration for the judge LLM (Autorater). - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. "A String", ], - }, + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. + "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. + "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. + "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. + }, + }, + "name": "A String", # Identifier. The resource name of the EvaluationMetric. Format: `projects/{project}/locations/{location}/evaluationMetrics/{evaluation_metric}` + "updateTime": "A String", # Output only. The time when the EvaluationMetric was last updated. +} + + evaluationMetricId: string, Optional. The ID to use for the EvaluationMetric, which will become the final component of the EvaluationMetric's resource name. This value should be 1-63 characters, and valid characters are /a-z-/. The first character must be a lowercase letter, and the last character must be a lowercase letter or number. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # EvaluationMetric is a resource that represents a reusable metric configuration. + "createTime": "A String", # Output only. The time when the EvaluationMetric was created. + "description": "A String", # Optional. A description of the EvaluationMetric. + "displayName": "A String", # Required. The user-friendly display name for the EvaluationMetric. + "encryptionSpec": { # Represents a customer-managed encryption key specification that can be applied to a Vertex AI resource. # Optional. Customer-managed encryption key spec for this EvaluationMetric. If set, this EvaluationMetric will be secured by this key. + "kmsKeyName": "A String", # Required. Resource name of the Cloud KMS key used to protect the resource. The Cloud KMS key must be in the same region as the resource. It must have the format `projects/{project}/locations/{location}/keyRings/{key_ring}/cryptoKeys/{crypto_key}`. + }, + "gcsUri": "A String", # Optional. The Google Cloud Storage URI that stores the metric specification.. + "labels": { # Optional. Labels for the evaluation metric. + "a_key": "A String", + }, + "metric": { # The metric used for running evaluations. # Optional. The metric configuration. Only LLMMetric and CustomCodeExecutionMetric are supported. + "aggregationMetrics": [ # Optional. The aggregation metrics to use. + "A String", + ], + "bleuSpec": { # Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. # Spec for bleu metric. + "useEffectiveOrder": True or False, # Optional. Whether to use_effective_order to compute bleu score. + }, + "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. + "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Required. The type of the computation based metric. + }, + "customCodeExecutionSpec": { # Specificies a metric that is populated by evaluating user-defined Python code. # Spec for Custom Code Execution metric. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used; requests from regions where the sandbox is unavailable will fail with UNIMPLEMENTED. + "evaluationFunction": "A String", # Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. + }, + "exactMatchSpec": { # Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. # Spec for exact match metric. + }, + "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. + "additionalConfig": { # Optional. Optional additional configuration for the metric. + "a_key": "", # Properties of the object. + }, + "judgeAutoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Optional configuration for the judge LLM (Autorater). + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, @@ -760,42 +1114,437 @@Method Details
"mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. + "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. + "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation + }, + }, + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. + "systemInstruction": "A String", # Optional. System instructions for the judge model. + }, + "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. + "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. + "a_key": "", # Properties of the object. }, - "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. - "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. - "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - }, + "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. + "description": "A String", # Optional. The description of the score explaining the directionality etc. + "max": 3.14, # Required. The maximum value of the score range (inclusive). + "min": 3.14, # Required. The minimum value of the score range (inclusive). + "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. + }, + "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. + "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. + "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. + "systemInstruction": "A String", # Optional. System instructions for pairwise metric. + }, + "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. + "systemInstruction": "A String", # Optional. System instructions for pointwise metric. + }, + "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -970,47 +1719,6 @@Method Details
}, "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - "systemInstruction": "A String", # Optional. System instructions for the judge model. - }, - "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. - "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. - "a_key": "", # Properties of the object. - }, - "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. - "description": "A String", # Optional. The description of the score explaining the directionality etc. - "max": 3.14, # Required. The maximum value of the score range (inclusive). - "min": 3.14, # Required. The minimum value of the score range (inclusive). - "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. - }, - "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. - }, - "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. - "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. - "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. - "systemInstruction": "A String", # Optional. System instructions for pairwise metric. - }, - "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. - "systemInstruction": "A String", # Optional. System instructions for pointwise metric. - }, - "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. }, }, "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. @@ -1248,45 +1956,222 @@Method Details
"mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, - }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. }, }, "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. @@ -1513,6 +2398,183 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. @@ -1760,6 +2822,183 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. @@ -1985,6 +3224,183 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationRuns.html b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationRuns.html index 34819228fe..11fd783610 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationRuns.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationRuns.html @@ -172,6 +172,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -562,6 +563,182 @@Method Details
"parameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. "metricResourceName": "A String", # Optional. Resource name of the metric definition. @@ -953,73 +1130,250 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, - }, - "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. - "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. - "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - }, - }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. "A String", ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. - }, - "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. - "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. - "compressionQuality": 42, # Optional. The compression quality of the output image. - "mimeType": "A String", # Optional. The image format that the output should be saved as. - }, - "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - "personGeneration": "A String", # Optional. Controls whether the model can generate people. - "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - }, - "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - "modelConfig": { # Config for model selection. # Optional. Config for model selection. - "featureSelectionPreference": "A String", # Required. Feature selection preference. - }, - "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. - { # Configuration for the model to configure output formatting and delivery. - "audio": { # Configuration for audio-specific output formatting. # Audio output format. - "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - "delivery": "A String", # Optional. Delivery mode for the generated content. - "mimeType": "A String", # Optional. The MIME type of the audio output. - "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. }, - "image": { # Configuration for image-specific output formatting. # Image output format. - "aspectRatio": "A String", # Optional. The aspect ratio for the image output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "imageSize": "A String", # Optional. The size of the image output. - "mimeType": "A String", # Optional. The MIME type of the image output. + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. + "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. + "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation + }, + }, + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. }, "text": { # Configuration for text-specific output formatting. # Text output format. "mimeType": "A String", # Optional. The IANA standard MIME type of the response. @@ -1178,235 +1532,220 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, - }, - "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. - "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. - "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. - }, - }, - "metricResourceName": "A String", # Optional. The resource name of the metric definition. - "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, - }, - "rubricBasedMetricSpec": { # Specification for a metric that is based on rubrics. # Spec for rubric based metric. - "inlineRubrics": { # Defines a list of rubrics, used when providing rubrics inline. # Use rubrics provided directly in the spec. - "rubrics": [ # The list of rubrics. - { # Message representing a single testable criterion for evaluation. One input prompt could have multiple rubrics. - "content": { # Content of the rubric, defining the testable criteria. # Required. The actual testable criteria for the rubric. - "property": { # Defines criteria based on a specific property. # Evaluation criteria based on a specific property. - "description": "A String", # Description of the property being evaluated. Example: "The model's response is grammatically correct." - }, - }, - "importance": "A String", # Optional. The relative importance of this rubric. - "rubricId": "A String", # Unique identifier for the rubric. This ID is used to refer to this rubric, e.g., in RubricVerdict. - "type": "A String", # Optional. A type designator for the rubric, which can inform how it's evaluated or interpreted by systems or users. It's recommended to use consistent, well-defined, upper snake_case strings. Examples: "SUMMARIZATION_QUALITY", "SAFETY_HARMFUL_CONTENT", "INSTRUCTION_ADHERENCE". - }, - ], - }, - "judgeAutoraterConfig": { # The autorater config used for the evaluation run. # Optional. Optional configuration for the judge LLM (Autorater). The definition of AutoraterConfig needs to be provided. - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. - "A String", - ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. - }, - "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. - "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. - "compressionQuality": 42, # Optional. The compression quality of the output image. - "mimeType": "A String", # Optional. The image format that the output should be saved as. - }, - "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - "personGeneration": "A String", # Optional. Controls whether the model can generate people. - "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - }, - "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - "modelConfig": { # Config for model selection. # Optional. Config for model selection. - "featureSelectionPreference": "A String", # Required. Feature selection preference. - }, - "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. - { # Configuration for the model to configure output formatting and delivery. - "audio": { # Configuration for audio-specific output formatting. # Audio output format. - "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - "delivery": "A String", # Optional. Delivery mode for the generated content. - "mimeType": "A String", # Optional. The MIME type of the audio output. - "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. - }, - "image": { # Configuration for image-specific output formatting. # Image output format. - "aspectRatio": "A String", # Optional. The aspect ratio for the image output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "imageSize": "A String", # Optional. The size of the image output. - "mimeType": "A String", # Optional. The MIME type of the image output. + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, - "text": { # Configuration for text-specific output formatting. # Text output format. - "mimeType": "A String", # Optional. The IANA standard MIME type of the response. - "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. }, - "video": { # Configuration for video-specific output formatting. # Video output format. - "aspectRatio": "A String", # The aspect ratio for the video output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "duration": "A String", # Optional. The duration for the video output. - "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. - "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. }, - }, - ], - "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. - "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - "A String", - ], - "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. - "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. - "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. - # Object with schema name: GoogleCloudAiplatformV1beta1Schema - ], - "default": "", # Optional. Default value to use if the field is not specified. - "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema - }, - "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. - "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` - "A String", - ], - "example": "", # Optional. Example of an instance of this schema. - "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. - "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. - "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. - "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. - "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. - "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. - "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. - "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. - "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. - "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. - "nullable": True or False, # Optional. Indicates if the value of this field can be null. - "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. - "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema - }, - "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. - "A String", - ], - "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring - "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. - "A String", - ], - "title": "A String", # Optional. Title for the schema. - "type": "A String", # Optional. Data type of the schema field. - }, - "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. - "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. - "modelRoutingPreference": "A String", # The model routing preference. - }, - "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. - "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. - }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. - }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". }, }, ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. - }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). - }, - }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, - }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "metricPromptTemplate": "A String", # Optional. Template for the prompt used by the judge model to evaluate against rubrics. - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics for evaluation using this specification. - "metricResourceName": "A String", # Optional. Resource name of the metric definition. - "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. - "A String", - ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. + "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. + "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. + "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. + }, + }, + "metricResourceName": "A String", # Optional. The resource name of the metric definition. + "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. @@ -1558,39 +1897,24 @@Method Details
}, "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # Optional. The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. }, - }, - ], - "outputConfig": { # The output config for the evaluation run. # Optional. The output config for the evaluation run. - "bigqueryDestination": { # The BigQuery location for the output content. # BigQuery destination for evaluation output. - "outputUri": "A String", # Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. - }, - "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud Storage destination for evaluation output. - "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. - }, - }, - "promptTemplate": { # Prompt template used for inference. # The prompt template used for inference. The values for variables in the prompt template are defined in EvaluationItem.EvaluationPrompt.PromptTemplateData.values. - "gcsUri": "A String", # Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". - "promptTemplate": "A String", # Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" - }, - "rubricConfigs": [ # Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. - { # Configuration for a rubric group to be generated/saved for evaluation. - "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. + "rubricBasedMetricSpec": { # Specification for a metric that is based on rubrics. # Spec for rubric based metric. + "inlineRubrics": { # Defines a list of rubrics, used when providing rubrics inline. # Use rubrics provided directly in the spec. + "rubrics": [ # The list of rubrics. + { # Message representing a single testable criterion for evaluation. One input prompt could have multiple rubrics. + "content": { # Content of the rubric, defining the testable criteria. # Required. The actual testable criteria for the rubric. + "property": { # Defines criteria based on a specific property. # Evaluation criteria based on a specific property. + "description": "A String", # Description of the property being evaluated. Example: "The model's response is grammatically correct." + }, + }, + "importance": "A String", # Optional. The relative importance of this rubric. + "rubricId": "A String", # Unique identifier for the rubric. This ID is used to refer to this rubric, e.g., in RubricVerdict. + "type": "A String", # Optional. A type designator for the rubric, which can inform how it's evaluated or interpreted by systems or users. It's recommended to use consistent, well-defined, upper snake_case strings. Examples: "SUMMARIZATION_QUALITY", "SAFETY_HARMFUL_CONTENT", "INSTRUCTION_ADHERENCE". + }, + ], }, - }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "metricResourceName": "A String", # Optional. Resource name of the metric definition. - "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "judgeAutoraterConfig": { # The autorater config used for the evaluation run. # Optional. Optional configuration for the judge LLM (Autorater). The definition of AutoraterConfig needs to be provided. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. @@ -1764,77 +2088,636 @@Method Details
}, "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # Optional. The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Required. The key used to save the generated rubrics. If a generation spec is provided, this key will be used for the name of the generated rubric group. Otherwise, this key will be used to look up the existing rubric group on the evaluation item. Note that if a rubric group key is specified on both a rubric config and an evaluation metric, the key from the metric will be used to select the rubrics for evaluation. - }, - ], - }, - "evaluationExperiment": "A String", # Optional. The resource name of the evaluation experiment for the evaluation run. Setting the evaluation experiment will add the run to the experiment. Multiple runs can be added to the same experiment. Format: `projects/{project}/locations/{location}/evaluationExperiments/{evaluation_experiment}` - "evaluationResults": { # The results of the evaluation run. # Output only. The results of the evaluation run. Only populated when the evaluation run's state is SUCCEEDED. - "evaluationSet": "A String", # The evaluation set where item level results are stored. - "lossAnalysisResults": [ # Output only. The results of the auto-loss analysis performed on the eval set associated with the run. - { # The top-level result for loss analysis, stored within an EvalSet. - "analysisTime": "A String", # The timestamp when this analysis was performed. - "clusters": [ # The list of identified loss clusters. - { # Represents a semantic grouping of failures (e.g., "Hallucination of Action"). - "clusterId": "A String", # Unique identifier for the loss cluster within the scope of the analysis result. - "examples": [ # A list of examples that belong to this cluster. This links the cluster back to the specific EvaluationItems and Rubrics. - { # Represents a specific example of a loss pattern. - "evaluationItem": "A String", # Reference to the persisted EvalItem resource name. Format: projects/.../locations/.../evaluationItems/{item_id} Used when analysis is run on an EvalSet. - "evaluationResult": { # Evaluation result. # The full evaluation result object provided inline. Used when the analysis is performed on ephemeral data (without an EvaluationSet). - "candidateResults": [ # Optional. The results for the metric. - { # Result for a single candidate. - "additionalResults": "", # Optional. Additional results for the metric. - "candidate": "A String", # Required. The candidate that is being evaluated. The value is the same as the candidate name in the EvaluationRequest. - "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # Output only. Error while evaluating the candidate for the metric. - "code": 42, # The status code, which should be an enum value of google.rpc.Code. - "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. - { - "a_key": "", # Properties of the object. Contains field @type with type URL. - }, - ], - "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. - }, - "explanation": "A String", # Optional. The explanation for the metric. - "metric": "A String", # Required. The metric that was evaluated. - "rubricVerdicts": [ # Optional. The rubric verdicts for the metric. - { # Represents the verdict of an evaluation against a single rubric. - "evaluatedRubric": { # Message representing a single testable criterion for evaluation. One input prompt could have multiple rubrics. # Required. The full rubric definition that was evaluated. Storing this ensures the verdict is self-contained and understandable, especially if the original rubric definition changes or was dynamically generated. - "content": { # Content of the rubric, defining the testable criteria. # Required. The actual testable criteria for the rubric. - "property": { # Defines criteria based on a specific property. # Evaluation criteria based on a specific property. - "description": "A String", # Description of the property being evaluated. Example: "The model's response is grammatically correct." - }, - }, - "importance": "A String", # Optional. The relative importance of this rubric. - "rubricId": "A String", # Unique identifier for the rubric. This ID is used to refer to this rubric, e.g., in RubricVerdict. - "type": "A String", # Optional. A type designator for the rubric, which can inform how it's evaluated or interpreted by systems or users. It's recommended to use consistent, well-defined, upper snake_case strings. Examples: "SUMMARIZATION_QUALITY", "SAFETY_HARMFUL_CONTENT", "INSTRUCTION_ADHERENCE". - }, - "reasoning": "A String", # Optional. Human-readable reasoning or explanation for the verdict. This can include specific examples or details from the evaluated content that justify the given verdict. - "verdict": True or False, # Required. Outcome of the evaluation against the rubric, represented as a boolean. `true` indicates a "Pass", `false` indicates a "Fail". - }, - ], - "score": 3.14, # Optional. The score for the metric. - }, + "metricPromptTemplate": "A String", # Optional. Template for the prompt used by the judge model to evaluate against rubrics. + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics for evaluation using this specification. + "metricResourceName": "A String", # Optional. Resource name of the metric definition. + "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", ], - "evaluationRequest": "A String", # Required. The request item that was evaluated. Format: projects/{project}/locations/{location}/evaluationItems/{evaluation_item} - "evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} - "metadata": "", # Optional. Metadata about the evaluation result. - "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. - "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. - { # Responses from model or agent. - "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. - "agents": { # Optional. A map containing the static configurations for each agent in the system. Key: agent_id (matches the `author` field in events). Value: The static configuration of the agent. - "a_key": { # Represents configuration for an Agent. - "agentId": "A String", # Required. Unique identifier of the agent. This ID is used to refer to this agent, e.g., in AgentEvent.author, or in the `sub_agents` field. It must be unique within the `agents` map. - "agentType": "A String", # Optional. The type or class of the agent (e.g., "LlmAgent", "RouterAgent", "ToolUseAgent"). Useful for the autorater to understand the expected behavior of the agent. - "description": "A String", # Optional. A high-level description of the agent's role and responsibilities. Critical for evaluating if the agent is routing tasks correctly. - "instruction": "A String", # Optional. Provides instructions for the LLM model, guiding the agent's behavior. Can be static or dynamic. Dynamic instructions can contain placeholders like {variable_name} that will be resolved at runtime using the `AgentEvent.state_delta` field. + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # Optional. The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], + }, + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. + }, + }, + ], + "outputConfig": { # The output config for the evaluation run. # Optional. The output config for the evaluation run. + "bigqueryDestination": { # The BigQuery location for the output content. # BigQuery destination for evaluation output. + "outputUri": "A String", # Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. + }, + "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud Storage destination for evaluation output. + "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. + }, + }, + "promptTemplate": { # Prompt template used for inference. # The prompt template used for inference. The values for variables in the prompt template are defined in EvaluationItem.EvaluationPrompt.PromptTemplateData.values. + "gcsUri": "A String", # Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". + "promptTemplate": "A String", # Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" + }, + "rubricConfigs": [ # Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. + { # Configuration for a rubric group to be generated/saved for evaluation. + "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "metricResourceName": "A String", # Optional. Resource name of the metric definition. + "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # Optional. The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], + }, + "rubricGroupKey": "A String", # Required. The key used to save the generated rubrics. If a generation spec is provided, this key will be used for the name of the generated rubric group. Otherwise, this key will be used to look up the existing rubric group on the evaluation item. Note that if a rubric group key is specified on both a rubric config and an evaluation metric, the key from the metric will be used to select the rubrics for evaluation. + }, + ], + }, + "evaluationExperiment": "A String", # Optional. The resource name of the evaluation experiment for the evaluation run. Setting the evaluation experiment will add the run to the experiment. Multiple runs can be added to the same experiment. Format: `projects/{project}/locations/{location}/evaluationExperiments/{evaluation_experiment}` + "evaluationResults": { # The results of the evaluation run. # Output only. The results of the evaluation run. Only populated when the evaluation run's state is SUCCEEDED. + "evaluationSet": "A String", # The evaluation set where item level results are stored. + "lossAnalysisResults": [ # Output only. The results of the auto-loss analysis performed on the eval set associated with the run. + { # The top-level result for loss analysis, stored within an EvalSet. + "analysisTime": "A String", # The timestamp when this analysis was performed. + "clusters": [ # The list of identified loss clusters. + { # Represents a semantic grouping of failures (e.g., "Hallucination of Action"). + "clusterId": "A String", # Unique identifier for the loss cluster within the scope of the analysis result. + "examples": [ # A list of examples that belong to this cluster. This links the cluster back to the specific EvaluationItems and Rubrics. + { # Represents a specific example of a loss pattern. + "evaluationItem": "A String", # Reference to the persisted EvalItem resource name. Format: projects/.../locations/.../evaluationItems/{item_id} Used when analysis is run on an EvalSet. + "evaluationResult": { # Evaluation result. # The full evaluation result object provided inline. Used when the analysis is performed on ephemeral data (without an EvaluationSet). + "candidateResults": [ # Optional. The results for the metric. + { # Result for a single candidate. + "additionalResults": "", # Optional. Additional results for the metric. + "candidate": "A String", # Required. The candidate that is being evaluated. The value is the same as the candidate name in the EvaluationRequest. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # Output only. Error while evaluating the candidate for the metric. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "explanation": "A String", # Optional. The explanation for the metric. + "metric": "A String", # Required. The metric that was evaluated. + "rubricVerdicts": [ # Optional. The rubric verdicts for the metric. + { # Represents the verdict of an evaluation against a single rubric. + "evaluatedRubric": { # Message representing a single testable criterion for evaluation. One input prompt could have multiple rubrics. # Required. The full rubric definition that was evaluated. Storing this ensures the verdict is self-contained and understandable, especially if the original rubric definition changes or was dynamically generated. + "content": { # Content of the rubric, defining the testable criteria. # Required. The actual testable criteria for the rubric. + "property": { # Defines criteria based on a specific property. # Evaluation criteria based on a specific property. + "description": "A String", # Description of the property being evaluated. Example: "The model's response is grammatically correct." + }, + }, + "importance": "A String", # Optional. The relative importance of this rubric. + "rubricId": "A String", # Unique identifier for the rubric. This ID is used to refer to this rubric, e.g., in RubricVerdict. + "type": "A String", # Optional. A type designator for the rubric, which can inform how it's evaluated or interpreted by systems or users. It's recommended to use consistent, well-defined, upper snake_case strings. Examples: "SUMMARIZATION_QUALITY", "SAFETY_HARMFUL_CONTENT", "INSTRUCTION_ADHERENCE". + }, + "reasoning": "A String", # Optional. Human-readable reasoning or explanation for the verdict. This can include specific examples or details from the evaluated content that justify the given verdict. + "verdict": True or False, # Required. Outcome of the evaluation against the rubric, represented as a boolean. `true` indicates a "Pass", `false` indicates a "Fail". + }, + ], + "score": 3.14, # Optional. The score for the metric. + }, + ], + "evaluationRequest": "A String", # Required. The request item that was evaluated. Format: projects/{project}/locations/{location}/evaluationItems/{evaluation_item} + "evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} + "metadata": "", # Optional. Metadata about the evaluation result. + "metric": "A String", # Required. The metric that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. + "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. + { # Responses from model or agent. + "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. + "agents": { # Optional. A map containing the static configurations for each agent in the system. Key: agent_id (matches the `author` field in events). Value: The static configuration of the agent. + "a_key": { # Represents configuration for an Agent. + "agentId": "A String", # Required. Unique identifier of the agent. This ID is used to refer to this agent, e.g., in AgentEvent.author, or in the `sub_agents` field. It must be unique within the `agents` map. + "agentType": "A String", # Optional. The type or class of the agent (e.g., "LlmAgent", "RouterAgent", "ToolUseAgent"). Useful for the autorater to understand the expected behavior of the agent. + "description": "A String", # Optional. A high-level description of the agent's role and responsibilities. Critical for evaluating if the agent is routing tasks correctly. + "instruction": "A String", # Optional. Provides instructions for the LLM model, guiding the agent's behavior. Can be static or dynamic. Dynamic instructions can contain placeholders like {variable_name} that will be resolved at runtime using the `AgentEvent.state_delta` field. "subAgents": [ # Optional. The list of valid agent IDs that this agent can delegate to. This defines the directed edges in the multi-agent system graph topology. "A String", ], @@ -1842,13 +2725,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2090,13 +2973,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2559,13 +3442,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2807,13 +3690,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3275,13 +4158,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -3523,13 +4406,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4124,13 +5007,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -4568,13 +5451,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -5035,6 +5918,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -5160,276 +6044,660 @@Method Details
"modelName": "A String", # The name of the model to use. Only public LLM models are accepted. }, }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "cloudLoggingConfig": { # Specifies configuration for exporting evaluation results to Cloud Logging. # Optional. Configuration for exporting evaluation results to Cloud Logging. + "project": "A String", # Optional. Google Cloud project to write logs to. Defaults to the request project. + "resourceLabels": { # Optional. MonitoredResource labels to associate the log with. The backend will automatically inject project and location. + "a_key": "A String", + }, + "resourceType": "A String", # Optional. MonitoredResource type. Defaults to "global" if unspecified. + "tracingContext": { # Tracing context for Observability correlation. # Optional. Tracing context for the evaluation run. + "conversationId": "A String", # Optional. Unique identifier for a conversation (session thread), used to store and correlate messages within a conversation. The value corresponds to the `gen_ai.conversation.id` field in the the OpenTelemetry GenAI attributes. + "spanId": "A String", # Optional. ID of the Cloud Trace span associated with the current operation in which the log is being written. e.g., `7a2190356c3fc94b`. If a span is being evaluated, this field should be populated. + "traceId": "A String", # Optional. Trace ID being written to Cloud Trace in association with this log entry. e.g., `12345`, the numeric ID from the resource name. If a trace or span is being evaluated, this field should be populated. + }, + }, + "datasetCustomMetrics": [ # Optional. Specifications for custom dataset-level aggregations. + { # Defines a custom dataset-level aggregation. + "aggregationFunction": "A String", # Required. The Python code string containing the aggregation function. Expected function signature: `def aggregate(instances: list[dict[str, Any]]) -> dict[str, float]:` The `instances` argument is a list of dictionaries, where each dictionary represents a single evaluation result item. The structure of each dictionary corresponds to the fields in the `EvaluationResult` message. This includes: - `"request"`: Contains the original input data and model inputs (from `EvaluationResult.EvaluationRequest`). - `"candidate_results"`: Contains the results of any instance-level metrics (from `EvaluationResult.CandidateResults`). Example of a single item in the `instances` list: { "request": { "prompt": {"text": "What is the capital of France?"}, "golden_response": {"text": "Paris"}, "candidate_responses": [{"candidate": "model-v1", "text": "Paris"}] }, "candidate_results": [ {"metric": "exact_match", "score": 1.0}, {"metric": "bleu", "score": 0.9} ] } + "displayName": "A String", # Optional. A display name for this custom summary metric. Used to prefix keys in the output summaryMetrics map. If not provided, a default name like "dataset_custom_metric_1", "dataset_custom_metric_2", etc., will be generated based on the order in the repeated field. + }, + ], + "lossAnalysisConfig": [ # Optional. Specifications for loss analysis. Each config can be specified for one metric. + { # Configuration for the loss analysis job. + "candidate": "A String", # Required. The candidate model/agent to analyze (e.g., "gemini-3.0-pro"). This targets the specific CandidateResult within the EvaluationResult. + "metric": "A String", # Required. The metric to analyze (e.g., "tool_use_quality"). This filters the EvaluationItems in the EvalSet to only those where EvaluationResult.metric matches this value. + }, + ], + "metrics": [ # Optional. The metrics to be calculated in the evaluation run. Required when analysis_configs is not set. + { # The metric used for evaluation runs. + "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. + "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Required. The type of the computation based metric. + }, + "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. + "additionalConfig": { # Optional. Optional additional configuration for the metric. + "a_key": "", # Properties of the object. + }, + "judgeAutoraterConfig": { # The autorater config used for the evaluation run. # Optional. Optional configuration for the judge LLM (Autorater). + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. "voiceName": "A String", # The name of the prebuilt voice to use. }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). - }, - }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, - }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "cloudLoggingConfig": { # Specifies configuration for exporting evaluation results to Cloud Logging. # Optional. Configuration for exporting evaluation results to Cloud Logging. - "project": "A String", # Optional. Google Cloud project to write logs to. Defaults to the request project. - "resourceLabels": { # Optional. MonitoredResource labels to associate the log with. The backend will automatically inject project and location. - "a_key": "A String", - }, - "resourceType": "A String", # Optional. MonitoredResource type. Defaults to "global" if unspecified. - "tracingContext": { # Tracing context for Observability correlation. # Optional. Tracing context for the evaluation run. - "conversationId": "A String", # Optional. Unique identifier for a conversation (session thread), used to store and correlate messages within a conversation. The value corresponds to the `gen_ai.conversation.id` field in the the OpenTelemetry GenAI attributes. - "spanId": "A String", # Optional. ID of the Cloud Trace span associated with the current operation in which the log is being written. e.g., `7a2190356c3fc94b`. If a span is being evaluated, this field should be populated. - "traceId": "A String", # Optional. Trace ID being written to Cloud Trace in association with this log entry. e.g., `12345`, the numeric ID from the resource name. If a trace or span is being evaluated, this field should be populated. - }, - }, - "datasetCustomMetrics": [ # Optional. Specifications for custom dataset-level aggregations. - { # Defines a custom dataset-level aggregation. - "aggregationFunction": "A String", # Required. The Python code string containing the aggregation function. Expected function signature: `def aggregate(instances: list[dict[str, Any]]) -> dict[str, float]:` The `instances` argument is a list of dictionaries, where each dictionary represents a single evaluation result item. The structure of each dictionary corresponds to the fields in the `EvaluationResult` message. This includes: - `"request"`: Contains the original input data and model inputs (from `EvaluationResult.EvaluationRequest`). - `"candidate_results"`: Contains the results of any instance-level metrics (from `EvaluationResult.CandidateResults`). Example of a single item in the `instances` list: { "request": { "prompt": {"text": "What is the capital of France?"}, "golden_response": {"text": "Paris"}, "candidate_responses": [{"candidate": "model-v1", "text": "Paris"}] }, "candidate_results": [ {"metric": "exact_match", "score": 1.0}, {"metric": "bleu", "score": 0.9} ] } - "displayName": "A String", # Optional. A display name for this custom summary metric. Used to prefix keys in the output summaryMetrics map. If not provided, a default name like "dataset_custom_metric_1", "dataset_custom_metric_2", etc., will be generated based on the order in the repeated field. - }, - ], - "lossAnalysisConfig": [ # Optional. Specifications for loss analysis. Each config can be specified for one metric. - { # Configuration for the loss analysis job. - "candidate": "A String", # Required. The candidate model/agent to analyze (e.g., "gemini-3.0-pro"). This targets the specific CandidateResult within the EvaluationResult. - "metric": "A String", # Required. The metric to analyze (e.g., "tool_use_quality"). This filters the EvaluationItems in the EvalSet to only those where EvaluationResult.metric matches this value. - }, - ], - "metrics": [ # Optional. The metrics to be calculated in the evaluation run. Required when analysis_configs is not set. - { # The metric used for evaluation runs. - "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. - "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - "a_key": "", # Properties of the object. - }, - "type": "A String", # Required. The type of the computation based metric. - }, - "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. - "additionalConfig": { # Optional. Optional additional configuration for the metric. - "a_key": "", # Properties of the object. }, - "judgeAutoraterConfig": { # The autorater config used for the evaluation run. # Optional. Optional configuration for the judge LLM (Autorater). - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "metricResourceName": "A String", # Optional. Resource name of the metric definition. + "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. "A String", ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. - }, - "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. - "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. - "compressionQuality": 42, # Optional. The compression quality of the output image. - "mimeType": "A String", # Optional. The image format that the output should be saved as. - }, - "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - "personGeneration": "A String", # Optional. Controls whether the model can generate people. - "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - }, - "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - "modelConfig": { # Config for model selection. # Optional. Config for model selection. - "featureSelectionPreference": "A String", # Required. Feature selection preference. - }, - "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. - { # Configuration for the model to configure output formatting and delivery. - "audio": { # Configuration for audio-specific output formatting. # Audio output format. - "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - "delivery": "A String", # Optional. Delivery mode for the generated content. - "mimeType": "A String", # Optional. The MIME type of the audio output. - "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. - }, - "image": { # Configuration for image-specific output formatting. # Image output format. - "aspectRatio": "A String", # Optional. The aspect ratio for the image output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "imageSize": "A String", # Optional. The size of the image output. - "mimeType": "A String", # Optional. The MIME type of the image output. - }, - "text": { # Configuration for text-specific output formatting. # Text output format. - "mimeType": "A String", # Optional. The IANA standard MIME type of the response. - "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. }, - "video": { # Configuration for video-specific output formatting. # Video output format. - "aspectRatio": "A String", # The aspect ratio for the video output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "duration": "A String", # Optional. The duration for the video output. - "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. - "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, - ], - "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. - "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - "A String", - ], - "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. - "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. - "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. - # Object with schema name: GoogleCloudAiplatformV1beta1Schema - ], - "default": "", # Optional. Default value to use if the field is not specified. - "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. }, - "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. - "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` - "A String", - ], - "example": "", # Optional. Example of an instance of this schema. - "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. - "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. - "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. - "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. - "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. - "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. - "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. - "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. - "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. - "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. - "nullable": True or False, # Optional. Indicates if the value of this field can be null. - "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. - "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. }, - "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. - "A String", + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, ], - "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring - "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - "title": "A String", # Optional. Title for the schema. - "type": "A String", # Optional. Data type of the schema field. - }, - "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. - "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. - "modelRoutingPreference": "A String", # The model routing preference. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. }, - "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. - "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. + ], }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # Optional. The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], + }, + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. + "systemInstruction": "A String", # Optional. System instructions for the judge model. + }, + "metric": "A String", # Required. The name of the metric. + "metricConfig": { # The metric used for running evaluations. # The metric config. + "aggregationMetrics": [ # Optional. The aggregation metrics to use. + "A String", + ], + "bleuSpec": { # Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. # Spec for bleu metric. + "useEffectiveOrder": True or False, # Optional. Whether to use_effective_order to compute bleu score. + }, + "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. + "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Required. The type of the computation based metric. + }, + "customCodeExecutionSpec": { # Specificies a metric that is populated by evaluating user-defined Python code. # Spec for Custom Code Execution metric. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used; requests from regions where the sandbox is unavailable will fail with UNIMPLEMENTED. + "evaluationFunction": "A String", # Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "exactMatchSpec": { # Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. # Spec for exact match metric. + }, + "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. + "additionalConfig": { # Optional. Optional additional configuration for the metric. "a_key": "", # Properties of the object. }, - }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "metricResourceName": "A String", # Optional. Resource name of the metric definition. - "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "judgeAutoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Optional configuration for the judge LLM (Autorater). "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). @@ -5600,231 +6868,418 @@Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # Optional. The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - "systemInstruction": "A String", # Optional. System instructions for the judge model. - }, - "metric": "A String", # Required. The name of the metric. - "metricConfig": { # The metric used for running evaluations. # The metric config. - "aggregationMetrics": [ # Optional. The aggregation metrics to use. - "A String", - ], - "bleuSpec": { # Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. # Spec for bleu metric. - "useEffectiveOrder": True or False, # Optional. Whether to use_effective_order to compute bleu score. - }, - "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. - "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - "a_key": "", # Properties of the object. - }, - "type": "A String", # Required. The type of the computation based metric. - }, - "customCodeExecutionSpec": { # Specificies a metric that is populated by evaluating user-defined Python code. # Spec for Custom Code Execution metric. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used; requests from regions where the sandbox is unavailable will fail with UNIMPLEMENTED. - "evaluationFunction": "A String", # Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. - }, - "exactMatchSpec": { # Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. # Spec for exact match metric. - }, - "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. - "additionalConfig": { # Optional. Optional additional configuration for the metric. - "a_key": "", # Properties of the object. + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "judgeAutoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Optional configuration for the judge LLM (Autorater). - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. - }, - "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. - "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. - "compressionQuality": 42, # Optional. The compression quality of the output image. - "mimeType": "A String", # Optional. The image format that the output should be saved as. - }, - "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - "personGeneration": "A String", # Optional. Controls whether the model can generate people. - "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - }, - "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - "modelConfig": { # Config for model selection. # Optional. Config for model selection. - "featureSelectionPreference": "A String", # Required. Feature selection preference. - }, - "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. - { # Configuration for the model to configure output formatting and delivery. - "audio": { # Configuration for audio-specific output formatting. # Audio output format. - "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - "delivery": "A String", # Optional. Delivery mode for the generated content. - "mimeType": "A String", # Optional. The MIME type of the audio output. - "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, }, - "image": { # Configuration for image-specific output formatting. # Image output format. - "aspectRatio": "A String", # Optional. The aspect ratio for the image output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "imageSize": "A String", # Optional. The size of the image output. - "mimeType": "A String", # Optional. The MIME type of the image output. + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. }, - "text": { # Configuration for text-specific output formatting. # Text output format. - "mimeType": "A String", # Optional. The IANA standard MIME type of the response. - "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, - "video": { # Configuration for video-specific output formatting. # Video output format. - "aspectRatio": "A String", # The aspect ratio for the video output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "duration": "A String", # Optional. The duration for the video output. - "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. - "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. + "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. + "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation + }, + }, + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, - ], - "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. - "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - "A String", - ], - "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. - "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. - "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. - # Object with schema name: GoogleCloudAiplatformV1beta1Schema - ], - "default": "", # Optional. Default value to use if the field is not specified. - "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. }, - "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. - "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` - "A String", - ], - "example": "", # Optional. Example of an instance of this schema. - "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. - "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. - "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. - "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. - "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. - "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. - "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. - "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. - "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. - "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. - "nullable": True or False, # Optional. Indicates if the value of this field can be null. - "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. - "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. }, - "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. - "A String", + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, ], - "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring - "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - "title": "A String", # Optional. Title for the schema. - "type": "A String", # Optional. Data type of the schema field. - }, - "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. - "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. - "modelRoutingPreference": "A String", # The model routing preference. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. }, - "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. - "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. + ], }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. + "systemInstruction": "A String", # Optional. System instructions for the judge model. + }, + "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. + "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. + "a_key": "", # Properties of the object. }, - "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. - "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. - "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - }, + "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. + "description": "A String", # Optional. The description of the score explaining the directionality etc. + "max": 3.14, # Required. The maximum value of the score range (inclusive). + "min": 3.14, # Required. The minimum value of the score range (inclusive). + "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. + }, + "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. + "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. + "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. + "systemInstruction": "A String", # Optional. System instructions for pairwise metric. + }, + "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. + "systemInstruction": "A String", # Optional. System instructions for pointwise metric. + }, + "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -5978,82 +7433,217 @@Method Details
"mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). - }, - }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. + "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. + "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. + "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. + }, + }, + "metricResourceName": "A String", # Optional. The resource name of the metric definition. + "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, }, - "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - "systemInstruction": "A String", # Optional. System instructions for the judge model. - }, - "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. - "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. - "a_key": "", # Properties of the object. - }, - "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. - "description": "A String", # Optional. The description of the score explaining the directionality etc. - "max": 3.14, # Required. The maximum value of the score range (inclusive). - "min": 3.14, # Required. The minimum value of the score range (inclusive). - "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. - }, - "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. - }, - "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. - "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. - "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. - "systemInstruction": "A String", # Optional. System instructions for pairwise metric. - }, - "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. - "systemInstruction": "A String", # Optional. System instructions for pointwise metric. - }, - "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, }, - "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. - "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. - "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. - }, - }, - "metricResourceName": "A String", # Optional. The resource name of the metric definition. - "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, }, "rubricBasedMetricSpec": { # Specification for a metric that is based on rubrics. # Spec for rubric based metric. "inlineRubrics": { # Defines a list of rubrics, used when providing rubrics inline. # Use rubrics provided directly in the spec. @@ -6191,63 +7781,267 @@Method Details
"autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. "modelRoutingPreference": "A String", # The model routing preference. }, - "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. - "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Optional. Template for the prompt used by the judge model to evaluate against rubrics. + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics for evaluation using this specification. + "metricResourceName": "A String", # Optional. Resource name of the metric definition. + "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. + ], }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # Optional. The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Optional. Template for the prompt used by the judge model to evaluate against rubrics. - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics for evaluation using this specification. - "metricResourceName": "A String", # Optional. Resource name of the metric definition. - "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. + }, + }, + ], + "outputConfig": { # The output config for the evaluation run. # Optional. The output config for the evaluation run. + "bigqueryDestination": { # The BigQuery location for the output content. # BigQuery destination for evaluation output. + "outputUri": "A String", # Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. + }, + "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud Storage destination for evaluation output. + "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. + }, + }, + "promptTemplate": { # Prompt template used for inference. # The prompt template used for inference. The values for variables in the prompt template are defined in EvaluationItem.EvaluationPrompt.PromptTemplateData.values. + "gcsUri": "A String", # Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". + "promptTemplate": "A String", # Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" + }, + "rubricConfigs": [ # Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. + { # Configuration for a rubric group to be generated/saved for evaluation. + "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. @@ -6421,34 +8215,6 @@Method Details
}, "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # Optional. The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. - }, - }, - ], - "outputConfig": { # The output config for the evaluation run. # Optional. The output config for the evaluation run. - "bigqueryDestination": { # The BigQuery location for the output content. # BigQuery destination for evaluation output. - "outputUri": "A String", # Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. - }, - "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud Storage destination for evaluation output. - "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. - }, - }, - "promptTemplate": { # Prompt template used for inference. # The prompt template used for inference. The values for variables in the prompt template are defined in EvaluationItem.EvaluationPrompt.PromptTemplateData.values. - "gcsUri": "A String", # Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". - "promptTemplate": "A String", # Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" - }, - "rubricConfigs": [ # Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. - { # Configuration for a rubric group to be generated/saved for evaluation. - "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. }, }, "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. @@ -6688,7 +8454,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -6705,13 +8471,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -6953,13 +8719,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7422,13 +9188,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -7670,13 +9436,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8138,13 +9904,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8386,13 +10152,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -8987,13 +10753,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9431,13 +11197,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -9940,6 +11706,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -10064,277 +11831,661 @@Method Details
"manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "cloudLoggingConfig": { # Specifies configuration for exporting evaluation results to Cloud Logging. # Optional. Configuration for exporting evaluation results to Cloud Logging. + "project": "A String", # Optional. Google Cloud project to write logs to. Defaults to the request project. + "resourceLabels": { # Optional. MonitoredResource labels to associate the log with. The backend will automatically inject project and location. + "a_key": "A String", + }, + "resourceType": "A String", # Optional. MonitoredResource type. Defaults to "global" if unspecified. + "tracingContext": { # Tracing context for Observability correlation. # Optional. Tracing context for the evaluation run. + "conversationId": "A String", # Optional. Unique identifier for a conversation (session thread), used to store and correlate messages within a conversation. The value corresponds to the `gen_ai.conversation.id` field in the the OpenTelemetry GenAI attributes. + "spanId": "A String", # Optional. ID of the Cloud Trace span associated with the current operation in which the log is being written. e.g., `7a2190356c3fc94b`. If a span is being evaluated, this field should be populated. + "traceId": "A String", # Optional. Trace ID being written to Cloud Trace in association with this log entry. e.g., `12345`, the numeric ID from the resource name. If a trace or span is being evaluated, this field should be populated. + }, + }, + "datasetCustomMetrics": [ # Optional. Specifications for custom dataset-level aggregations. + { # Defines a custom dataset-level aggregation. + "aggregationFunction": "A String", # Required. The Python code string containing the aggregation function. Expected function signature: `def aggregate(instances: list[dict[str, Any]]) -> dict[str, float]:` The `instances` argument is a list of dictionaries, where each dictionary represents a single evaluation result item. The structure of each dictionary corresponds to the fields in the `EvaluationResult` message. This includes: - `"request"`: Contains the original input data and model inputs (from `EvaluationResult.EvaluationRequest`). - `"candidate_results"`: Contains the results of any instance-level metrics (from `EvaluationResult.CandidateResults`). Example of a single item in the `instances` list: { "request": { "prompt": {"text": "What is the capital of France?"}, "golden_response": {"text": "Paris"}, "candidate_responses": [{"candidate": "model-v1", "text": "Paris"}] }, "candidate_results": [ {"metric": "exact_match", "score": 1.0}, {"metric": "bleu", "score": 0.9} ] } + "displayName": "A String", # Optional. A display name for this custom summary metric. Used to prefix keys in the output summaryMetrics map. If not provided, a default name like "dataset_custom_metric_1", "dataset_custom_metric_2", etc., will be generated based on the order in the repeated field. + }, + ], + "lossAnalysisConfig": [ # Optional. Specifications for loss analysis. Each config can be specified for one metric. + { # Configuration for the loss analysis job. + "candidate": "A String", # Required. The candidate model/agent to analyze (e.g., "gemini-3.0-pro"). This targets the specific CandidateResult within the EvaluationResult. + "metric": "A String", # Required. The metric to analyze (e.g., "tool_use_quality"). This filters the EvaluationItems in the EvalSet to only those where EvaluationResult.metric matches this value. + }, + ], + "metrics": [ # Optional. The metrics to be calculated in the evaluation run. Required when analysis_configs is not set. + { # The metric used for evaluation runs. + "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. + "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Required. The type of the computation based metric. + }, + "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. + "additionalConfig": { # Optional. Optional additional configuration for the metric. + "a_key": "", # Properties of the object. + }, + "judgeAutoraterConfig": { # The autorater config used for the evaluation run. # Optional. Optional configuration for the judge LLM (Autorater). + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. "voiceName": "A String", # The name of the prebuilt voice to use. }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). - }, - }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, - }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "cloudLoggingConfig": { # Specifies configuration for exporting evaluation results to Cloud Logging. # Optional. Configuration for exporting evaluation results to Cloud Logging. - "project": "A String", # Optional. Google Cloud project to write logs to. Defaults to the request project. - "resourceLabels": { # Optional. MonitoredResource labels to associate the log with. The backend will automatically inject project and location. - "a_key": "A String", - }, - "resourceType": "A String", # Optional. MonitoredResource type. Defaults to "global" if unspecified. - "tracingContext": { # Tracing context for Observability correlation. # Optional. Tracing context for the evaluation run. - "conversationId": "A String", # Optional. Unique identifier for a conversation (session thread), used to store and correlate messages within a conversation. The value corresponds to the `gen_ai.conversation.id` field in the the OpenTelemetry GenAI attributes. - "spanId": "A String", # Optional. ID of the Cloud Trace span associated with the current operation in which the log is being written. e.g., `7a2190356c3fc94b`. If a span is being evaluated, this field should be populated. - "traceId": "A String", # Optional. Trace ID being written to Cloud Trace in association with this log entry. e.g., `12345`, the numeric ID from the resource name. If a trace or span is being evaluated, this field should be populated. - }, - }, - "datasetCustomMetrics": [ # Optional. Specifications for custom dataset-level aggregations. - { # Defines a custom dataset-level aggregation. - "aggregationFunction": "A String", # Required. The Python code string containing the aggregation function. Expected function signature: `def aggregate(instances: list[dict[str, Any]]) -> dict[str, float]:` The `instances` argument is a list of dictionaries, where each dictionary represents a single evaluation result item. The structure of each dictionary corresponds to the fields in the `EvaluationResult` message. This includes: - `"request"`: Contains the original input data and model inputs (from `EvaluationResult.EvaluationRequest`). - `"candidate_results"`: Contains the results of any instance-level metrics (from `EvaluationResult.CandidateResults`). Example of a single item in the `instances` list: { "request": { "prompt": {"text": "What is the capital of France?"}, "golden_response": {"text": "Paris"}, "candidate_responses": [{"candidate": "model-v1", "text": "Paris"}] }, "candidate_results": [ {"metric": "exact_match", "score": 1.0}, {"metric": "bleu", "score": 0.9} ] } - "displayName": "A String", # Optional. A display name for this custom summary metric. Used to prefix keys in the output summaryMetrics map. If not provided, a default name like "dataset_custom_metric_1", "dataset_custom_metric_2", etc., will be generated based on the order in the repeated field. - }, - ], - "lossAnalysisConfig": [ # Optional. Specifications for loss analysis. Each config can be specified for one metric. - { # Configuration for the loss analysis job. - "candidate": "A String", # Required. The candidate model/agent to analyze (e.g., "gemini-3.0-pro"). This targets the specific CandidateResult within the EvaluationResult. - "metric": "A String", # Required. The metric to analyze (e.g., "tool_use_quality"). This filters the EvaluationItems in the EvalSet to only those where EvaluationResult.metric matches this value. - }, - ], - "metrics": [ # Optional. The metrics to be calculated in the evaluation run. Required when analysis_configs is not set. - { # The metric used for evaluation runs. - "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. - "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - "a_key": "", # Properties of the object. - }, - "type": "A String", # Required. The type of the computation based metric. - }, - "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. - "additionalConfig": { # Optional. Optional additional configuration for the metric. - "a_key": "", # Properties of the object. }, - "judgeAutoraterConfig": { # The autorater config used for the evaluation run. # Optional. Optional configuration for the judge LLM (Autorater). - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "metricResourceName": "A String", # Optional. Resource name of the metric definition. + "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. "A String", ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. - }, - "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. - "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. - "compressionQuality": 42, # Optional. The compression quality of the output image. - "mimeType": "A String", # Optional. The image format that the output should be saved as. - }, - "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - "personGeneration": "A String", # Optional. Controls whether the model can generate people. - "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - }, - "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - "modelConfig": { # Config for model selection. # Optional. Config for model selection. - "featureSelectionPreference": "A String", # Required. Feature selection preference. - }, - "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. - { # Configuration for the model to configure output formatting and delivery. - "audio": { # Configuration for audio-specific output formatting. # Audio output format. - "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - "delivery": "A String", # Optional. Delivery mode for the generated content. - "mimeType": "A String", # Optional. The MIME type of the audio output. - "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. - }, - "image": { # Configuration for image-specific output formatting. # Image output format. - "aspectRatio": "A String", # Optional. The aspect ratio for the image output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "imageSize": "A String", # Optional. The size of the image output. - "mimeType": "A String", # Optional. The MIME type of the image output. - }, - "text": { # Configuration for text-specific output formatting. # Text output format. - "mimeType": "A String", # Optional. The IANA standard MIME type of the response. - "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. }, - "video": { # Configuration for video-specific output formatting. # Video output format. - "aspectRatio": "A String", # The aspect ratio for the video output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "duration": "A String", # Optional. The duration for the video output. - "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. - "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, - ], - "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. - "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - "A String", - ], - "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. - "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. - "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. - # Object with schema name: GoogleCloudAiplatformV1beta1Schema - ], - "default": "", # Optional. Default value to use if the field is not specified. - "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. }, - "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. - "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` - "A String", - ], - "example": "", # Optional. Example of an instance of this schema. - "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. - "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. - "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. - "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. - "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. - "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. - "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. - "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. - "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. - "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. - "nullable": True or False, # Optional. Indicates if the value of this field can be null. - "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. - "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. }, - "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. - "A String", + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, ], - "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring - "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - "title": "A String", # Optional. Title for the schema. - "type": "A String", # Optional. Data type of the schema field. - }, - "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. - "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. - "modelRoutingPreference": "A String", # The model routing preference. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. }, - "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. - "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. + ], }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # Optional. The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. + "systemInstruction": "A String", # Optional. System instructions for the judge model. + }, + "metric": "A String", # Required. The name of the metric. + "metricConfig": { # The metric used for running evaluations. # The metric config. + "aggregationMetrics": [ # Optional. The aggregation metrics to use. + "A String", + ], + "bleuSpec": { # Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. # Spec for bleu metric. + "useEffectiveOrder": True or False, # Optional. Whether to use_effective_order to compute bleu score. + }, + "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. + "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. + "a_key": "", # Properties of the object. + }, + "type": "A String", # Required. The type of the computation based metric. + }, + "customCodeExecutionSpec": { # Specificies a metric that is populated by evaluating user-defined Python code. # Spec for Custom Code Execution metric. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used; requests from regions where the sandbox is unavailable will fail with UNIMPLEMENTED. + "evaluationFunction": "A String", # Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. + }, + "exactMatchSpec": { # Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. # Spec for exact match metric. + }, + "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. + "additionalConfig": { # Optional. Optional additional configuration for the metric. "a_key": "", # Properties of the object. }, - }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "metricResourceName": "A String", # Optional. Resource name of the metric definition. - "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "judgeAutoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Optional configuration for the judge LLM (Autorater). "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). @@ -10505,231 +12656,418 @@Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # Optional. The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - "systemInstruction": "A String", # Optional. System instructions for the judge model. - }, - "metric": "A String", # Required. The name of the metric. - "metricConfig": { # The metric used for running evaluations. # The metric config. - "aggregationMetrics": [ # Optional. The aggregation metrics to use. - "A String", - ], - "bleuSpec": { # Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. # Spec for bleu metric. - "useEffectiveOrder": True or False, # Optional. Whether to use_effective_order to compute bleu score. - }, - "computationBasedMetricSpec": { # Specification for a computation based metric. # Spec for a computation based metric. - "parameters": { # Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - "a_key": "", # Properties of the object. - }, - "type": "A String", # Required. The type of the computation based metric. - }, - "customCodeExecutionSpec": { # Specificies a metric that is populated by evaluating user-defined Python code. # Spec for Custom Code Execution metric. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used; requests from regions where the sandbox is unavailable will fail with UNIMPLEMENTED. - "evaluationFunction": "A String", # Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. - }, - "exactMatchSpec": { # Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. # Spec for exact match metric. - }, - "llmBasedMetricSpec": { # Specification for an LLM based metric. # Spec for an LLM based metric. - "additionalConfig": { # Optional. Optional additional configuration for the metric. - "a_key": "", # Properties of the object. + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "judgeAutoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Optional configuration for the judge LLM (Autorater). - "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. - "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. - "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). - "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - "A String", - ], - "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - "A String", - ], - "diarization": True or False, # Optional. Configures speaker diarization. - "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. - }, - "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - "A String", - ], - "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. - "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - }, - "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. - }, - "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. - "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. - "compressionQuality": 42, # Optional. The compression quality of the output image. - "mimeType": "A String", # Optional. The image format that the output should be saved as. - }, - "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - "personGeneration": "A String", # Optional. Controls whether the model can generate people. - "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - }, - "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - "modelConfig": { # Config for model selection. # Optional. Config for model selection. - "featureSelectionPreference": "A String", # Required. Feature selection preference. - }, - "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. - { # Configuration for the model to configure output formatting and delivery. - "audio": { # Configuration for audio-specific output formatting. # Audio output format. - "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - "delivery": "A String", # Optional. Delivery mode for the generated content. - "mimeType": "A String", # Optional. The MIME type of the audio output. - "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. }, - "image": { # Configuration for image-specific output formatting. # Image output format. - "aspectRatio": "A String", # Optional. The aspect ratio for the image output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "imageSize": "A String", # Optional. The size of the image output. - "mimeType": "A String", # Optional. The MIME type of the image output. + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, - "text": { # Configuration for text-specific output formatting. # Text output format. - "mimeType": "A String", # Optional. The IANA standard MIME type of the response. - "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. + "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. + "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation + }, + }, + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. }, - "video": { # Configuration for video-specific output formatting. # Video output format. - "aspectRatio": "A String", # The aspect ratio for the video output. - "delivery": "A String", # Optional. Delivery mode for the generated content. - "duration": "A String", # Optional. The duration for the video output. - "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. - "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. }, - ], - "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. - "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - "A String", - ], - "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. - "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. - "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. - # Object with schema name: GoogleCloudAiplatformV1beta1Schema - ], - "default": "", # Optional. Default value to use if the field is not specified. - "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. }, - "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. - "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` - "A String", - ], - "example": "", # Optional. Example of an instance of this schema. - "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. - "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. - "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. - "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. - "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. - "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. - "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. - "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. - "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. - "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. - "nullable": True or False, # Optional. Indicates if the value of this field can be null. - "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. - "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. - "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. }, - "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. - "A String", + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, ], - "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring - "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. "A String", ], - "title": "A String", # Optional. Title for the schema. - "type": "A String", # Optional. Data type of the schema field. - }, - "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. - "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. - "modelRoutingPreference": "A String", # The model routing preference. + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. }, - "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. - "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, }, - }, - "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. - "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. - "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. - "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - { # Configuration for a single speaker in a multi-speaker setup. - "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. - }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - }, - ], - }, - "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. - "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. - "voiceName": "A String", # The name of the prebuilt voice to use. + ], }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. + "systemInstruction": "A String", # Optional. System instructions for the judge model. + }, + "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. + "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. + "a_key": "", # Properties of the object. }, - "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. - "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. - "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - }, + "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. + "description": "A String", # Optional. The description of the score explaining the directionality etc. + "max": 3.14, # Required. The maximum value of the score range (inclusive). + "min": 3.14, # Required. The minimum value of the score range (inclusive). + "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. + }, + "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. + "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. + "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. + "systemInstruction": "A String", # Optional. System instructions for pairwise metric. + }, + "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. + "systemInstruction": "A String", # Optional. System instructions for pointwise metric. + }, + "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -10902,63 +13240,198 @@Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. + "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. + "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. + "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. + }, + }, + "metricResourceName": "A String", # Optional. The resource name of the metric definition. + "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, }, - "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - "systemInstruction": "A String", # Optional. System instructions for the judge model. - }, - "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. - "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. - "a_key": "", # Properties of the object. - }, - "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. - "description": "A String", # Optional. The description of the score explaining the directionality etc. - "max": 3.14, # Required. The maximum value of the score range (inclusive). - "min": 3.14, # Required. The minimum value of the score range (inclusive). - "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. - }, - "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. - }, - "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. - "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. - "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. - "systemInstruction": "A String", # Optional. System instructions for pairwise metric. - }, - "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. - "systemInstruction": "A String", # Optional. System instructions for pointwise metric. - }, - "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, }, - "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. - "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. - "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. - }, - }, - "metricResourceName": "A String", # Optional. The resource name of the metric definition. - "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, }, "rubricBasedMetricSpec": { # Specification for a metric that is based on rubrics. # Spec for rubric based metric. "inlineRubrics": { # Defines a list of rubrics, used when providing rubrics inline. # Use rubrics provided directly in the spec. @@ -11131,28 +13604,232 @@Method Details
"voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Optional. Template for the prompt used by the judge model to evaluate against rubrics. + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics for evaluation using this specification. + "metricResourceName": "A String", # Optional. Resource name of the metric definition. + "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # Optional. The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Optional. Template for the prompt used by the judge model to evaluate against rubrics. - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics for evaluation using this specification. - "metricResourceName": "A String", # Optional. Resource name of the metric definition. - "modelConfig": { # The autorater config used for the evaluation run. # Optional. Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. + }, + }, + ], + "outputConfig": { # The output config for the evaluation run. # Optional. The output config for the evaluation run. + "bigqueryDestination": { # The BigQuery location for the output content. # BigQuery destination for evaluation output. + "outputUri": "A String", # Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. + }, + "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud Storage destination for evaluation output. + "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. + }, + }, + "promptTemplate": { # Prompt template used for inference. # The prompt template used for inference. The values for variables in the prompt template are defined in EvaluationItem.EvaluationPrompt.PromptTemplateData.values. + "gcsUri": "A String", # Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". + "promptTemplate": "A String", # Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" + }, + "rubricConfigs": [ # Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. + { # Configuration for a rubric group to be generated/saved for evaluation. + "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. @@ -11326,34 +14003,6 @@Method Details
}, "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "promptTemplate": "A String", # Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # Optional. The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. - }, - }, - ], - "outputConfig": { # The output config for the evaluation run. # Optional. The output config for the evaluation run. - "bigqueryDestination": { # The BigQuery location for the output content. # BigQuery destination for evaluation output. - "outputUri": "A String", # Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. - }, - "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud Storage destination for evaluation output. - "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. - }, - }, - "promptTemplate": { # Prompt template used for inference. # The prompt template used for inference. The values for variables in the prompt template are defined in EvaluationItem.EvaluationPrompt.PromptTemplateData.values. - "gcsUri": "A String", # Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". - "promptTemplate": "A String", # Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" - }, - "rubricConfigs": [ # Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. - { # Configuration for a rubric group to be generated/saved for evaluation. - "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. }, }, "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. @@ -11593,7 +14242,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -11610,13 +14259,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -11858,13 +14507,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12327,13 +14976,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -12575,13 +15224,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13043,13 +15692,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13291,13 +15940,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -13892,13 +16541,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14336,13 +16985,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -14816,6 +17465,7 @@Method Details
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. }, "evaluationConfig": { # The Evalution configuration used for the evaluation run. # Optional. The configuration used for the evaluation. Optional when analysis_configs is provided. + "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. "autoraterConfig": { # The autorater config used for the evaluation run. # Optional. The autorater config for the evaluation run. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -15198,14 +17848,190 @@Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # Specification for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. "metricResourceName": "A String", # Optional. Resource name of the metric definition. @@ -15566,46 +18392,441 @@Method Details
"prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. "voiceName": "A String", # The name of the prebuilt voice to use. }, - "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. - "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. + "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. + "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. + "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. + "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation + }, + }, + "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. + "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). }, - "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - "A String", - ], - "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. - "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. - }, - "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - "translationConfig": { # Config for translation features. # Optional. Config for translation. - "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. + "rubricContentType": "A String", # The type of rubric content to be generated. + "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. + "A String", + ], }, - "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model. - "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, + "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. + "systemInstruction": "A String", # Optional. System instructions for the judge model. + }, + "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. + "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. + "a_key": "", # Properties of the object. }, - "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. - "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. - "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used. - "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - }, + "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. + "description": "A String", # Optional. The description of the score explaining the directionality etc. + "max": 3.14, # Required. The maximum value of the score range (inclusive). + "min": 3.14, # Required. The minimum value of the score range (inclusive). + "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. }, - "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. - "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation. + "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. + }, + "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. + "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. + "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. + "systemInstruction": "A String", # Optional. System instructions for pairwise metric. + }, + "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. + "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. + "returnRawOutput": True or False, # Optional. Whether to return raw output. + }, + "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. + "systemInstruction": "A String", # Optional. System instructions for pointwise metric. + }, + "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. @@ -15778,63 +18999,198 @@Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). }, }, - "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, + }, + "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. + "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. + "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. + "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. + }, + }, + "metricResourceName": "A String", # Optional. The resource name of the metric definition. + "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. + "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". + "parameters": { # Optional. The parameters needed to run the pre-defined metric. + "a_key": "", # Properties of the object. + }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, }, - "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - "rubricContentType": "A String", # The type of rubric content to be generated. - "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - "A String", - ], - }, - "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - "systemInstruction": "A String", # Optional. System instructions for the judge model. - }, - "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization. - "otherMetadata": { # Optional. Flexible metadata for user-defined attributes. - "a_key": "", # Properties of the object. - }, - "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting. - "description": "A String", # Optional. The description of the score explaining the directionality etc. - "max": 3.14, # Required. The maximum value of the score range (inclusive). - "min": 3.14, # Required. The minimum value of the score range (inclusive). - "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. - }, - "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. - }, - "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric. - "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response. - "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric. - "systemInstruction": "A String", # Optional. System instructions for pairwise metric. - }, - "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric. - "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty. - "returnRawOutput": True or False, # Optional. Whether to return raw output. - }, - "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric. - "systemInstruction": "A String", # Optional. System instructions for pointwise metric. - }, - "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. }, }, - "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. - "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. - "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. - }, - }, - "metricResourceName": "A String", # Optional. The resource name of the metric definition. - "predefinedMetricSpec": { # Specification for a pre-defined metric. # Spec for a pre-defined metric. - "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - "parameters": { # Optional. The parameters needed to run the pre-defined metric. - "a_key": "", # Properties of the object. - }, }, "rubricBasedMetricSpec": { # Specification for a metric that is based on rubrics. # Spec for rubric based metric. "inlineRubrics": { # Defines a list of rubrics, used when providing rubrics inline. # Use rubrics provided directly in the spec. @@ -16231,6 +19587,182 @@Method Details
"parameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names: "intent_extraction", "rubric_generation", "rubric_validation". + "a_key": { # The autorater config used for the evaluation run. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "sampleCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification. "metricResourceName": "A String", # Optional. Resource name of the metric definition. @@ -16469,7 +20001,7 @@Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run} "metadata": "", # Optional. Metadata about the evaluation result. "metric": "A String", # Required. The metric that was evaluated. - "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated. + "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated. "candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison. { # Responses from model or agent. "agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation. @@ -16486,13 +20018,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -16734,13 +20266,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -17203,13 +20735,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -17451,13 +20983,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -17919,13 +21451,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -18167,13 +21699,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -18768,13 +22300,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -19212,13 +22744,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationSets.html b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationSets.html index a7f74c1981..c186cef830 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationSets.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.evaluationSets.html @@ -132,13 +132,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -407,13 +407,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -724,13 +724,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1033,13 +1033,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1375,13 +1375,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1669,13 +1669,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -1945,13 +1945,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.html b/docs/dyn/aiplatform_v1beta1.projects.locations.html index 188703cdea..d89ddb43b4 100644 --- a/docs/dyn/aiplatform_v1beta1.projects.locations.html +++ b/docs/dyn/aiplatform_v1beta1.projects.locations.html @@ -481,13 +481,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -795,13 +795,13 @@Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval). "codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation. }, - "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. - "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use. + "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations. + "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool. "A String", ], - "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request. - "environment": "A String", # Required. The environment being operated. - "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. + "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request. + "environment": "A String", # Required. The target environment where the computer use tool operates. + "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. "A String", ], }, @@ -2232,6 +2232,183 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, + "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT. + "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. + "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` + "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. + "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs. + "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. + "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition). + "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. + "A String", + ], + "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. + "A String", + ], + "diarization": True or False, # Optional. Configures speaker diarization. + "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically. + }, + "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. + "A String", + ], + "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio. + "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. + "A String", + ], + }, + "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. + "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation. + }, + "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. + "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. + "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. + "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead. + "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" + "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images. + "compressionQuality": 42, # Optional. The compression quality of the output image. + "mimeType": "A String", # Optional. The image format that the output should be saved as. + }, + "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. + "personGeneration": "A String", # Optional. Controls whether the model can generate people. + "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. + }, + "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. + "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. + "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. + "modelConfig": { # Config for model selection. # Optional. Config for model selection. + "featureSelectionPreference": "A String", # Required. Feature selection preference. + }, + "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. + "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery. + { # Configuration for the model to configure output formatting and delivery. + "audio": { # Configuration for audio-specific output formatting. # Audio output format. + "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). + "delivery": "A String", # Optional. Delivery mode for the generated content. + "mimeType": "A String", # Optional. The MIME type of the audio output. + "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz. + }, + "image": { # Configuration for image-specific output formatting. # Image output format. + "aspectRatio": "A String", # Optional. The aspect ratio for the image output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "imageSize": "A String", # Optional. The size of the image output. + "mimeType": "A String", # Optional. The MIME type of the image output. + }, + "text": { # Configuration for text-specific output formatting. # Text output format. + "mimeType": "A String", # Optional. The IANA standard MIME type of the response. + "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON. + }, + "video": { # Configuration for video-specific output formatting. # Video output format. + "aspectRatio": "A String", # The aspect ratio for the video output. + "delivery": "A String", # Optional. Delivery mode for the generated content. + "duration": "A String", # Optional. The duration for the video output. + "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. + "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". + }, + }, + ], + "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead. + "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. + "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. + "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. + "A String", + ], + "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead. + "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema. + "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. + # Object with schema name: GoogleCloudAiplatformV1beta1Schema + ], + "default": "", # Optional. Default value to use if the field is not specified. + "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. + "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` + "A String", + ], + "example": "", # Optional. Example of an instance of this schema. + "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. + "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array. + "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. + "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. + "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. + "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. + "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. + "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. + "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. + "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. + "nullable": True or False, # Optional. Indicates if the value of this field can be null. + "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. + "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. + "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema + }, + "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. + "A String", + ], + "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring + "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. + "A String", + ], + "title": "A String", # Optional. Title for the schema. + "type": "A String", # Optional. Data type of the schema field. + }, + "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration. + "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request. + "modelRoutingPreference": "A String", # The model routing preference. + }, + "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually. + "modelName": "A String", # The name of the model to use. Only public LLM models are accepted. + }, + }, + "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. + "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config. + "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis. + "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`. + "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. + { # Configuration for a single speaker in a multi-speaker setup. + "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt. + "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + ], + }, + "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use. + "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice. + "voiceName": "A String", # The name of the prebuilt voice to use. + }, + "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample. + "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. + "voiceSampleAudio": "A String", # Optional. The sample of the custom voice. + }, + "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). + }, + }, + "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. + "A String", + ], + "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. + "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking. + "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. + "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. + "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate. + }, + "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. + "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. + "translationConfig": { # Config for translation features. # Optional. Config for translation. + "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. + "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). + }, + }, + "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. + }, + }, }, "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result. "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response. @@ -2457,87 +2634,265 @@Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric. "a_key": "", # Properties of the object. }, - }, - "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric. - "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum. - "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score. - }, - }, - ], - "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output. - "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output. - "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. - }, - }, -} - - x__xgafv: string, V1 error format. - Allowed values - 1 - v1 error format - 2 - v2 error format - -Returns: - An object of the form: - - { # This resource represents a long-running operation that is the result of a network API call. - "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. - "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. - "code": 42, # The status code, which should be an enum value of google.rpc.Code. - "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. - { - "a_key": "", # Properties of the object. Contains field @type with type URL. - }, - ], - "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. - }, - "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. - "a_key": "", # Properties of the object. Contains field @type with type URL. - }, - "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. - "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. - "a_key": "", # Properties of the object. Contains field @type with type URL. - }, -}
evaluateInstances(location, body=None, x__xgafv=None)
- Evaluates instances based on a given metric.
-
-Args:
- location: string, Required. The resource name of the Location to evaluate the instances. Format: `projects/{project}/locations/{location}` (required)
- body: object, The request body.
- The object takes the form of:
-
-{ # Request message for EvaluationService.EvaluateInstances.
- "autoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Autorater config used for evaluation. Not applicable for predefined metrics (PredefinedMetricSpec); the server uses its own model configuration for predefined metrics and this field is ignored.
- "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
- "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
- "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
- "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
- "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
- "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
- "A String",
- ],
- "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
- "A String",
- ],
- "diarization": True or False, # Optional. Configures speaker diarization.
- "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
- },
- "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
- "A String",
- ],
- "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
- "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
- "A String",
- ],
- },
- "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
- "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
- },
- "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
+ "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
+ "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
+ "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
+ },
+ },
+ ],
+ "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
+ "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
+ "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
+ },
+ },
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # This resource represents a long-running operation that is the result of a network API call.
+ "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available.
+ "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation.
+ "code": 42, # The status code, which should be an enum value of google.rpc.Code.
+ "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use.
+ {
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ ],
+ "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.
+ },
+ "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`.
+ "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+}
+evaluateInstances(location, body=None, x__xgafv=None)
+ Evaluates instances based on a given metric.
+
+Args:
+ location: string, Required. The resource name of the Location to evaluate the instances. Format: `projects/{project}/locations/{location}` (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request message for EvaluationService.EvaluateInstances.
+ "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the request location is provided elsewhere in the request, this flag must be set to true or the request will fail.
+ "autoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Autorater config used for evaluation. Not applicable for predefined metrics (PredefinedMetricSpec); the server uses its own model configuration for predefined metrics and this field is ignored.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
"enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
"frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
"imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
@@ -2868,13 +3223,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -3223,13 +3578,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -3669,13 +4024,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -3917,13 +4272,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -4275,13 +4630,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -4523,13 +4878,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -5501,21 +5856,416 @@ Method Details
},
"samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
- },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
+ "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
+ "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
+ "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
+ },
+ },
+ "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
+ "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
+ "rubricContentType": "A String", # The type of rubric content to be generated.
+ "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
+ "A String",
+ ],
+ },
+ "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
+ "systemInstruction": "A String", # Optional. System instructions for the judge model.
+ },
+ "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
+ "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
+ "a_key": "", # Properties of the object.
+ },
+ "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
+ "description": "A String", # Optional. The description of the score explaining the directionality etc.
+ "max": 3.14, # Required. The maximum value of the score range (inclusive).
+ "min": 3.14, # Required. The minimum value of the score range (inclusive).
+ "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
},
- "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
- "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
- "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
- "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
- },
+ "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
+ },
+ "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
+ "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
+ "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
},
- "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
- "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
+ },
+ "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
+ },
+ "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
"autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
"flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
"generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
@@ -5690,47 +6440,6 @@ Method Details
},
"samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
- "rubricContentType": "A String", # The type of rubric content to be generated.
- "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
- "A String",
- ],
- },
- "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
- "systemInstruction": "A String", # Optional. System instructions for the judge model.
- },
- "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
- "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
- "a_key": "", # Properties of the object.
- },
- "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
- "description": "A String", # Optional. The description of the score explaining the directionality etc.
- "max": 3.14, # Required. The maximum value of the score range (inclusive).
- "min": 3.14, # Required. The minimum value of the score range (inclusive).
- "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
- },
- "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
- },
- "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
- "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
- "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
- },
- "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
- },
- "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
@@ -5947,6 +6656,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -6172,6 +7058,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
"rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
@@ -7104,13 +8167,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -7448,6 +8511,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rubricGenerationSpec": { # Specification for how rubrics should be generated. # Optional. Specification for how the rubrics should be generated.
"modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
@@ -7715,7 +8955,7 @@ Method Details
"evaluationRun": "A String", # Required. The evaluation run that was used to generate the result. Format: projects/{project}/locations/{location}/evaluationRuns/{evaluation_run}
"metadata": "", # Optional. Metadata about the evaluation result.
"metric": "A String", # Required. The metric that was evaluated.
- "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. # Required. The request that was evaluated.
+ "request": { # A single evaluation request supporting input for both single-turn model generation and multi-turn agent execution traces. Valid input modes: 1. Inference Mode: `prompt` is set (containing text or AgentData context). 2. Offline Eval Mode: `prompt` is unset, and `candidate_responses` contains `agent_data` (the completed execution trace). Validation Rule: Either `prompt` must be set, OR at least one of the `candidate_responses` must contain `agent_data`. Metrics receive each candidate as an `EvaluationInstance`, whose fields are the names metrics can reference: `prompt` -> `prompt`, the candidate -> `response` (or `agent_eval_data` for `agent_data`), `golden_response` -> `reference`, `rubrics` -> `rubric_groups`, and `prompt.prompt_template_data.values` and `extra_context` -> `other_data`. # Required. The request that was evaluated.
"candidateResponses": [ # Optional. Responses from model under test and other baseline models for comparison.
{ # Responses from model or agent.
"agentData": { # Represents data specific to multi-turn agent evaluations. # Optional. Represents the complete execution trace of a multi-turn conversation, which can involve single or multiple agents. This field is used to provide the full output of an agent's run, including all turns and events, for direct evaluation.
@@ -7732,13 +8972,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -7980,13 +9220,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -8449,13 +9689,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -8697,13 +9937,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -9165,13 +10405,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -9413,13 +10653,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -10179,13 +11419,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.onlineEvaluators.html b/docs/dyn/aiplatform_v1beta1.projects.locations.onlineEvaluators.html
index 8885e31d80..41893c41cf 100644
--- a/docs/dyn/aiplatform_v1beta1.projects.locations.onlineEvaluators.html
+++ b/docs/dyn/aiplatform_v1beta1.projects.locations.onlineEvaluators.html
@@ -436,6 +436,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -661,6 +838,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
"rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
@@ -993,45 +1347,222 @@ Method Details
"mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
"voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
},
- "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
},
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
},
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
],
- },
- "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
- "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
- "voiceName": "A String", # The name of the prebuilt voice to use.
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
},
- "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
- "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
- "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
},
- "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
},
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
- "A String",
- ],
- "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
- "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
- "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
- "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
- "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
- },
- "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
- "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
- "translationConfig": { # Config for translation features. # Optional. Config for translation.
- "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
- "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
- },
- },
- "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
- },
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
@@ -1258,6 +1789,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
"rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
@@ -1566,23 +2274,418 @@ Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
},
},
- "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
+ "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
+ "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
+ "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
+ },
+ },
+ "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
+ "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
+ "rubricContentType": "A String", # The type of rubric content to be generated.
+ "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
+ "A String",
+ ],
},
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
- },
+ "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
+ "systemInstruction": "A String", # Optional. System instructions for the judge model.
+ },
+ "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
+ "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
+ "a_key": "", # Properties of the object.
},
- "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
- "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
- "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
- "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
- },
+ "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
+ "description": "A String", # Optional. The description of the score explaining the directionality etc.
+ "max": 3.14, # Required. The maximum value of the score range (inclusive).
+ "min": 3.14, # Required. The minimum value of the score range (inclusive).
+ "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
},
- "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
- "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
+ },
+ "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
+ "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
+ "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
+ },
+ "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
+ },
+ "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
"autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
"flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
"generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
@@ -1757,47 +2860,6 @@ Method Details
},
"samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
- "rubricContentType": "A String", # The type of rubric content to be generated.
- "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
- "A String",
- ],
- },
- "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
- "systemInstruction": "A String", # Optional. System instructions for the judge model.
- },
- "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
- "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
- "a_key": "", # Properties of the object.
- },
- "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
- "description": "A String", # Optional. The description of the score explaining the directionality etc.
- "max": 3.14, # Required. The maximum value of the score range (inclusive).
- "min": 3.14, # Required. The minimum value of the score range (inclusive).
- "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
- },
- "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
- },
- "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
- "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
- "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
- },
- "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
- },
- "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
@@ -2079,45 +3141,222 @@ Method Details
"mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
"voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
},
- "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
},
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
},
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
],
- },
- "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
- "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
- "voiceName": "A String", # The name of the prebuilt voice to use.
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
},
- "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
- "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
- "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
},
- "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
},
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
- "A String",
- ],
- "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
- "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
- "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
- "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
- "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
- },
- "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
- "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
- "translationConfig": { # Config for translation features. # Optional. Config for translation.
- "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
- "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
- },
- },
- "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
- },
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
@@ -2344,6 +3583,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
"rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.publishers.models.html b/docs/dyn/aiplatform_v1beta1.projects.locations.publishers.models.html
index 923f38961d..19ed6148a1 100644
--- a/docs/dyn/aiplatform_v1beta1.projects.locations.publishers.models.html
+++ b/docs/dyn/aiplatform_v1beta1.projects.locations.publishers.models.html
@@ -646,13 +646,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -1603,13 +1603,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -2974,13 +2974,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.semanticGovernancePolicies.html b/docs/dyn/aiplatform_v1beta1.projects.locations.semanticGovernancePolicies.html
index 1243865fae..6e4f4af97e 100644
--- a/docs/dyn/aiplatform_v1beta1.projects.locations.semanticGovernancePolicies.html
+++ b/docs/dyn/aiplatform_v1beta1.projects.locations.semanticGovernancePolicies.html
@@ -116,7 +116,7 @@ Method Details
The object takes the form of:
{ # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent.
- "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy.
+ "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}`
"agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID`
"agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request.
"denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters.
@@ -129,7 +129,7 @@ Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy.
{ # Represents a governance policy applied to MCP tools used by an Agent.
"mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`
- "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer.
+ "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified.
"A String",
],
},
@@ -220,7 +220,7 @@ Method Details
An object of the form:
{ # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent.
- "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy.
+ "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}`
"agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID`
"agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request.
"denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters.
@@ -233,7 +233,7 @@ Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy.
{ # Represents a governance policy applied to MCP tools used by an Agent.
"mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`
- "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer.
+ "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified.
"A String",
],
},
@@ -264,7 +264,7 @@ Method Details
"nextPageToken": "A String", # A token to retrieve the next page of results. Pass to ListSemanticGovernancePoliciesRequest.page_token to obtain that page.
"semanticGovernancePolicies": [ # The list of SemanticGovernancePolicies.
{ # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent.
- "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy.
+ "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}`
"agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID`
"agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request.
"denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters.
@@ -277,7 +277,7 @@ Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy.
{ # Represents a governance policy applied to MCP tools used by an Agent.
"mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`
- "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer.
+ "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified.
"A String",
],
},
@@ -314,7 +314,7 @@ Method Details
The object takes the form of:
{ # Represents a governance policy applied to a specific Agent and optionally a specific Tool within that Agent.
- "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy.
+ "agent": "A String", # Required. The name of the agent in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/agents/{agent}`
"agentIdentity": "A String", # Output only. Represents the principal of the agent, used by the Policy Decision Point (PDP) for governance checks. For more information, see https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity Format: `principal://TRUST_DOMAIN/NAMESPACE/AGENT_NAME` Example: `principal://agents.global.org-ORGANIZATION_ID.system.id.goog/resources/aiplatform/projects/PROJECT_NUMBER/locations/LOCATION/reasoningEngines/AGENT_ENGINE_ID`
"agentResponseCustomization": { # Customizes the agent's response to the end user when a `SemanticGovernancePolicy` is evaluated (for example, with a custom message shown on denial). # Optional. Settings for customizing the agent's response to end users when this policy is evaluated, such as messages displayed when the policy denies a request.
"denialMessage": "A String", # Optional. Custom message shown to the end user when the policy check results in a denial. Use this to explain the rationale to the user. Max 1000 characters.
@@ -327,7 +327,7 @@ Method Details
"mcpTools": [ # Optional. The McpTools that are affected by this policy.
{ # Represents a governance policy applied to MCP tools used by an Agent.
"mcpServer": "A String", # Required. The resource name of the McpServer in Agent Registry that is affected by this policy. Format: `projects/{project}/locations/{location}/mcpServers/{mcp_server}`
- "tools": [ # Optional. The resource names of the McpTools used by the Agent that is affected by this policy. If not specified, the policy applies to all McpTools in the McpServer.
+ "tools": [ # Required. The resource names of the tools used by the Agent that is affected by this policy. Currently, exactly one tool must be specified.
"A String",
],
},
diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.tuningJobs.html b/docs/dyn/aiplatform_v1beta1.projects.locations.tuningJobs.html
index a22a34b188..9f44fb1b94 100644
--- a/docs/dyn/aiplatform_v1beta1.projects.locations.tuningJobs.html
+++ b/docs/dyn/aiplatform_v1beta1.projects.locations.tuningJobs.html
@@ -1318,6 +1318,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -1543,87 +1720,264 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
- },
- "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
- "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
- "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
- "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
- },
- },
- ],
- "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
- "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
- "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
- },
- },
- },
- "exportLastCheckpointOnly": True or False, # Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.
- "hyperParameters": { # Hyperparameters for SFT. # Optional. Hyperparameters for SFT.
- "adapterSize": "A String", # Optional. Adapter size for tuning.
- "batchSize": "A String", # Optional. Batch size for tuning. This feature is only available for open source models.
- "epochCount": "A String", # Optional. Number of complete passes the model makes over the entire training dataset during training.
- "learningRate": 3.14, # Optional. Learning rate for tuning. Mutually exclusive with `learning_rate_multiplier`. This feature is only available for open source models.
- "learningRateMultiplier": 3.14, # Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with `learning_rate`. This feature is only available for 1P models.
- },
- "trainingDatasetUri": "A String", # Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
- "tuningMode": "A String", # Tuning mode.
- "validationDatasetUri": "A String", # Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
- },
- "tunedModel": { # The Model Registry Model and Online Prediction Endpoint associated with this TuningJob. # Output only. The tuned model resources associated with this TuningJob.
- "checkpoints": [ # Output only. The checkpoints associated with this TunedModel. This field is only populated for tuning jobs that enable intermediate checkpoints.
- { # TunedModelCheckpoint for the Tuned Model of a Tuning Job.
- "checkpointId": "A String", # The ID of the checkpoint.
- "endpoint": "A String", # The Endpoint resource name that the checkpoint is deployed to. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
- "epoch": "A String", # The epoch of the checkpoint.
- "step": "A String", # The step of the checkpoint.
- },
- ],
- "endpoint": "A String", # Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
- "model": "A String", # Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}@{version_id}` When tuning from a base model, the version ID will be 1. For continuous tuning, if the provided tuned_model_display_name is set and different from parent model's display name, the tuned model will have a new parent model with version 1. Otherwise the version id will be incremented by 1 from the last version ID in the parent model. E.g., `projects/{project}/locations/{location}/models/{model}@{last_version_id + 1}`
- },
- "tunedModelDisplayName": "A String", # Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
- "tuningDataStats": { # The tuning data statistic values for TuningJob. # Output only. The tuning data statistics associated with this TuningJob.
- "distillationDataStats": { # Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model. # Output only. Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model.
- "trainingDatasetStats": { # Statistics computed over a tuning dataset. # Output only. Statistics computed for the training dataset.
- "contentsPerExampleDistribution": { # Distribution computed over a tuning dataset. # Output only. Dataset distributions for the number of contents per example.
- "buckets": [ # Output only. Defines the histogram bucket.
- { # Dataset bucket used to create a histogram for the distribution given a population of values.
- "count": "A String", # Output only. Number of values in the bucket.
- "left": 3.14, # Output only. Left bound of the bucket.
- "right": 3.14, # Output only. Right bound of the bucket.
- },
- ],
- "max": 3.14, # Output only. The maximum of the population values.
- "mean": 3.14, # Output only. The arithmetic mean of the values in the population.
- "median": 3.14, # Output only. The median of the values in the population.
- "min": 3.14, # Output only. The minimum of the population values.
- "p5": 3.14, # Output only. The 5th percentile of the values in the population.
- "p95": 3.14, # Output only. The 95th percentile of the values in the population.
- "sum": 3.14, # Output only. Sum of a given population of values.
- },
- "droppedExampleIndices": [ # Output only. A partial sample of the indices (starting from 1) of the dropped examples.
- "A String",
- ],
- "droppedExampleReasons": [ # Output only. For each index in `dropped_example_indices`, the user-facing reason why the example was dropped.
- "A String",
- ],
- "reinforcementTuningUserDatasetExamples": { # Sample reinforcement tuning user data in the training dataset. The contents are truncated for better UI showing. # Output only. Sample user dataset examples in the training dataset uri for Reinforcement Tuning.
- "userDatasetExamples": [ # List of user datasset examples showing to user.
- { # User-facing format for Gemini Reinforcement Tuning examples on Vertex.
- "contents": [ # Multi-turn contents that represents the Prompt.
- { # The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message.
- "parts": [ # Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part.
- { # A datatype containing media that is part of a multi-part Content message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. For media types that are not text, `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.
- "audioTranscription": { # The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. # Optional. Audio (input or output) transcription. This is only set when this `Part` contains audio data.
- "speakerLabel": "A String", # Optional. A label identifying the speaker of this audio segment (e.g. `spk_1`, `spk_2`). Present when `diarization` is set.
- "text": "A String", # Required. The transcription text of this audio segment.
- "words": [ # Optional. Detailed word-level transcriptions and timing details. Present when `word_timestamp` is set.
- { # Information about a single recognized word.
- "endOffset": "A String", # Optional. End offset in time of the word relative to the start of the audio.
- "startOffset": "A String", # Optional. Start offset in time of the word relative to the start of the audio.
- "word": "A String", # Required. Transcript of the word.
- },
- ],
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
+ "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
+ "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
+ "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
+ },
+ },
+ ],
+ "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
+ "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
+ "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
+ },
+ },
+ },
+ "exportLastCheckpointOnly": True or False, # Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.
+ "hyperParameters": { # Hyperparameters for SFT. # Optional. Hyperparameters for SFT.
+ "adapterSize": "A String", # Optional. Adapter size for tuning.
+ "batchSize": "A String", # Optional. Batch size for tuning. This feature is only available for open source models.
+ "epochCount": "A String", # Optional. Number of complete passes the model makes over the entire training dataset during training.
+ "learningRate": 3.14, # Optional. Learning rate for tuning. Mutually exclusive with `learning_rate_multiplier`. This feature is only available for open source models.
+ "learningRateMultiplier": 3.14, # Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with `learning_rate`. This feature is only available for 1P models.
+ },
+ "trainingDatasetUri": "A String", # Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
+ "tuningMode": "A String", # Tuning mode.
+ "validationDatasetUri": "A String", # Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
+ },
+ "tunedModel": { # The Model Registry Model and Online Prediction Endpoint associated with this TuningJob. # Output only. The tuned model resources associated with this TuningJob.
+ "checkpoints": [ # Output only. The checkpoints associated with this TunedModel. This field is only populated for tuning jobs that enable intermediate checkpoints.
+ { # TunedModelCheckpoint for the Tuned Model of a Tuning Job.
+ "checkpointId": "A String", # The ID of the checkpoint.
+ "endpoint": "A String", # The Endpoint resource name that the checkpoint is deployed to. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "epoch": "A String", # The epoch of the checkpoint.
+ "step": "A String", # The step of the checkpoint.
+ },
+ ],
+ "endpoint": "A String", # Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "model": "A String", # Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}@{version_id}` When tuning from a base model, the version ID will be 1. For continuous tuning, if the provided tuned_model_display_name is set and different from parent model's display name, the tuned model will have a new parent model with version 1. Otherwise the version id will be incremented by 1 from the last version ID in the parent model. E.g., `projects/{project}/locations/{location}/models/{model}@{last_version_id + 1}`
+ },
+ "tunedModelDisplayName": "A String", # Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
+ "tuningDataStats": { # The tuning data statistic values for TuningJob. # Output only. The tuning data statistics associated with this TuningJob.
+ "distillationDataStats": { # Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model. # Output only. Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model.
+ "trainingDatasetStats": { # Statistics computed over a tuning dataset. # Output only. Statistics computed for the training dataset.
+ "contentsPerExampleDistribution": { # Distribution computed over a tuning dataset. # Output only. Dataset distributions for the number of contents per example.
+ "buckets": [ # Output only. Defines the histogram bucket.
+ { # Dataset bucket used to create a histogram for the distribution given a population of values.
+ "count": "A String", # Output only. Number of values in the bucket.
+ "left": 3.14, # Output only. Left bound of the bucket.
+ "right": 3.14, # Output only. Right bound of the bucket.
+ },
+ ],
+ "max": 3.14, # Output only. The maximum of the population values.
+ "mean": 3.14, # Output only. The arithmetic mean of the values in the population.
+ "median": 3.14, # Output only. The median of the values in the population.
+ "min": 3.14, # Output only. The minimum of the population values.
+ "p5": 3.14, # Output only. The 5th percentile of the values in the population.
+ "p95": 3.14, # Output only. The 95th percentile of the values in the population.
+ "sum": 3.14, # Output only. Sum of a given population of values.
+ },
+ "droppedExampleIndices": [ # Output only. A partial sample of the indices (starting from 1) of the dropped examples.
+ "A String",
+ ],
+ "droppedExampleReasons": [ # Output only. For each index in `dropped_example_indices`, the user-facing reason why the example was dropped.
+ "A String",
+ ],
+ "reinforcementTuningUserDatasetExamples": { # Sample reinforcement tuning user data in the training dataset. The contents are truncated for better UI showing. # Output only. Sample user dataset examples in the training dataset uri for Reinforcement Tuning.
+ "userDatasetExamples": [ # List of user datasset examples showing to user.
+ { # User-facing format for Gemini Reinforcement Tuning examples on Vertex.
+ "contents": [ # Multi-turn contents that represents the Prompt.
+ { # The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message.
+ "parts": [ # Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part.
+ { # A datatype containing media that is part of a multi-part Content message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. For media types that are not text, `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.
+ "audioTranscription": { # The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. # Optional. Audio (input or output) transcription. This is only set when this `Part` contains audio data.
+ "speakerLabel": "A String", # Optional. A label identifying the speaker of this audio segment (e.g. `spk_1`, `spk_2`). Present when `diarization` is set.
+ "text": "A String", # Required. The transcription text of this audio segment.
+ "words": [ # Optional. Detailed word-level transcriptions and timing details. Present when `word_timestamp` is set.
+ { # Information about a single recognized word.
+ "endOffset": "A String", # Optional. End offset in time of the word relative to the start of the audio.
+ "startOffset": "A String", # Optional. Start offset in time of the word relative to the start of the audio.
+ "word": "A String", # Required. Transcript of the word.
+ },
+ ],
},
"codeExecutionResult": { # Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. # Optional. The result of executing the ExecutableCode.
"id": "A String", # Optional. The identifier of the `ExecutableCode` part this result is for. Only populated if the corresponding `ExecutableCode` has an id.
@@ -1807,13 +2161,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -2696,13 +3050,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -4460,6 +4814,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -4685,87 +5216,264 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
- },
- "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
- "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
- "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
- "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
- },
- },
- ],
- "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
- "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
- "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
- },
- },
- },
- "exportLastCheckpointOnly": True or False, # Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.
- "hyperParameters": { # Hyperparameters for SFT. # Optional. Hyperparameters for SFT.
- "adapterSize": "A String", # Optional. Adapter size for tuning.
- "batchSize": "A String", # Optional. Batch size for tuning. This feature is only available for open source models.
- "epochCount": "A String", # Optional. Number of complete passes the model makes over the entire training dataset during training.
- "learningRate": 3.14, # Optional. Learning rate for tuning. Mutually exclusive with `learning_rate_multiplier`. This feature is only available for open source models.
- "learningRateMultiplier": 3.14, # Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with `learning_rate`. This feature is only available for 1P models.
- },
- "trainingDatasetUri": "A String", # Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
- "tuningMode": "A String", # Tuning mode.
- "validationDatasetUri": "A String", # Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
- },
- "tunedModel": { # The Model Registry Model and Online Prediction Endpoint associated with this TuningJob. # Output only. The tuned model resources associated with this TuningJob.
- "checkpoints": [ # Output only. The checkpoints associated with this TunedModel. This field is only populated for tuning jobs that enable intermediate checkpoints.
- { # TunedModelCheckpoint for the Tuned Model of a Tuning Job.
- "checkpointId": "A String", # The ID of the checkpoint.
- "endpoint": "A String", # The Endpoint resource name that the checkpoint is deployed to. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
- "epoch": "A String", # The epoch of the checkpoint.
- "step": "A String", # The step of the checkpoint.
- },
- ],
- "endpoint": "A String", # Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
- "model": "A String", # Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}@{version_id}` When tuning from a base model, the version ID will be 1. For continuous tuning, if the provided tuned_model_display_name is set and different from parent model's display name, the tuned model will have a new parent model with version 1. Otherwise the version id will be incremented by 1 from the last version ID in the parent model. E.g., `projects/{project}/locations/{location}/models/{model}@{last_version_id + 1}`
- },
- "tunedModelDisplayName": "A String", # Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
- "tuningDataStats": { # The tuning data statistic values for TuningJob. # Output only. The tuning data statistics associated with this TuningJob.
- "distillationDataStats": { # Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model. # Output only. Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model.
- "trainingDatasetStats": { # Statistics computed over a tuning dataset. # Output only. Statistics computed for the training dataset.
- "contentsPerExampleDistribution": { # Distribution computed over a tuning dataset. # Output only. Dataset distributions for the number of contents per example.
- "buckets": [ # Output only. Defines the histogram bucket.
- { # Dataset bucket used to create a histogram for the distribution given a population of values.
- "count": "A String", # Output only. Number of values in the bucket.
- "left": 3.14, # Output only. Left bound of the bucket.
- "right": 3.14, # Output only. Right bound of the bucket.
- },
- ],
- "max": 3.14, # Output only. The maximum of the population values.
- "mean": 3.14, # Output only. The arithmetic mean of the values in the population.
- "median": 3.14, # Output only. The median of the values in the population.
- "min": 3.14, # Output only. The minimum of the population values.
- "p5": 3.14, # Output only. The 5th percentile of the values in the population.
- "p95": 3.14, # Output only. The 95th percentile of the values in the population.
- "sum": 3.14, # Output only. Sum of a given population of values.
- },
- "droppedExampleIndices": [ # Output only. A partial sample of the indices (starting from 1) of the dropped examples.
- "A String",
- ],
- "droppedExampleReasons": [ # Output only. For each index in `dropped_example_indices`, the user-facing reason why the example was dropped.
- "A String",
- ],
- "reinforcementTuningUserDatasetExamples": { # Sample reinforcement tuning user data in the training dataset. The contents are truncated for better UI showing. # Output only. Sample user dataset examples in the training dataset uri for Reinforcement Tuning.
- "userDatasetExamples": [ # List of user datasset examples showing to user.
- { # User-facing format for Gemini Reinforcement Tuning examples on Vertex.
- "contents": [ # Multi-turn contents that represents the Prompt.
- { # The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message.
- "parts": [ # Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part.
- { # A datatype containing media that is part of a multi-part Content message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. For media types that are not text, `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.
- "audioTranscription": { # The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. # Optional. Audio (input or output) transcription. This is only set when this `Part` contains audio data.
- "speakerLabel": "A String", # Optional. A label identifying the speaker of this audio segment (e.g. `spk_1`, `spk_2`). Present when `diarization` is set.
- "text": "A String", # Required. The transcription text of this audio segment.
- "words": [ # Optional. Detailed word-level transcriptions and timing details. Present when `word_timestamp` is set.
- { # Information about a single recognized word.
- "endOffset": "A String", # Optional. End offset in time of the word relative to the start of the audio.
- "startOffset": "A String", # Optional. Start offset in time of the word relative to the start of the audio.
- "word": "A String", # Required. Transcript of the word.
- },
- ],
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
+ "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
+ "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
+ "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
+ },
+ },
+ ],
+ "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
+ "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
+ "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
+ },
+ },
+ },
+ "exportLastCheckpointOnly": True or False, # Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.
+ "hyperParameters": { # Hyperparameters for SFT. # Optional. Hyperparameters for SFT.
+ "adapterSize": "A String", # Optional. Adapter size for tuning.
+ "batchSize": "A String", # Optional. Batch size for tuning. This feature is only available for open source models.
+ "epochCount": "A String", # Optional. Number of complete passes the model makes over the entire training dataset during training.
+ "learningRate": 3.14, # Optional. Learning rate for tuning. Mutually exclusive with `learning_rate_multiplier`. This feature is only available for open source models.
+ "learningRateMultiplier": 3.14, # Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with `learning_rate`. This feature is only available for 1P models.
+ },
+ "trainingDatasetUri": "A String", # Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
+ "tuningMode": "A String", # Tuning mode.
+ "validationDatasetUri": "A String", # Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
+ },
+ "tunedModel": { # The Model Registry Model and Online Prediction Endpoint associated with this TuningJob. # Output only. The tuned model resources associated with this TuningJob.
+ "checkpoints": [ # Output only. The checkpoints associated with this TunedModel. This field is only populated for tuning jobs that enable intermediate checkpoints.
+ { # TunedModelCheckpoint for the Tuned Model of a Tuning Job.
+ "checkpointId": "A String", # The ID of the checkpoint.
+ "endpoint": "A String", # The Endpoint resource name that the checkpoint is deployed to. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "epoch": "A String", # The epoch of the checkpoint.
+ "step": "A String", # The step of the checkpoint.
+ },
+ ],
+ "endpoint": "A String", # Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "model": "A String", # Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}@{version_id}` When tuning from a base model, the version ID will be 1. For continuous tuning, if the provided tuned_model_display_name is set and different from parent model's display name, the tuned model will have a new parent model with version 1. Otherwise the version id will be incremented by 1 from the last version ID in the parent model. E.g., `projects/{project}/locations/{location}/models/{model}@{last_version_id + 1}`
+ },
+ "tunedModelDisplayName": "A String", # Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
+ "tuningDataStats": { # The tuning data statistic values for TuningJob. # Output only. The tuning data statistics associated with this TuningJob.
+ "distillationDataStats": { # Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model. # Output only. Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model.
+ "trainingDatasetStats": { # Statistics computed over a tuning dataset. # Output only. Statistics computed for the training dataset.
+ "contentsPerExampleDistribution": { # Distribution computed over a tuning dataset. # Output only. Dataset distributions for the number of contents per example.
+ "buckets": [ # Output only. Defines the histogram bucket.
+ { # Dataset bucket used to create a histogram for the distribution given a population of values.
+ "count": "A String", # Output only. Number of values in the bucket.
+ "left": 3.14, # Output only. Left bound of the bucket.
+ "right": 3.14, # Output only. Right bound of the bucket.
+ },
+ ],
+ "max": 3.14, # Output only. The maximum of the population values.
+ "mean": 3.14, # Output only. The arithmetic mean of the values in the population.
+ "median": 3.14, # Output only. The median of the values in the population.
+ "min": 3.14, # Output only. The minimum of the population values.
+ "p5": 3.14, # Output only. The 5th percentile of the values in the population.
+ "p95": 3.14, # Output only. The 95th percentile of the values in the population.
+ "sum": 3.14, # Output only. Sum of a given population of values.
+ },
+ "droppedExampleIndices": [ # Output only. A partial sample of the indices (starting from 1) of the dropped examples.
+ "A String",
+ ],
+ "droppedExampleReasons": [ # Output only. For each index in `dropped_example_indices`, the user-facing reason why the example was dropped.
+ "A String",
+ ],
+ "reinforcementTuningUserDatasetExamples": { # Sample reinforcement tuning user data in the training dataset. The contents are truncated for better UI showing. # Output only. Sample user dataset examples in the training dataset uri for Reinforcement Tuning.
+ "userDatasetExamples": [ # List of user datasset examples showing to user.
+ { # User-facing format for Gemini Reinforcement Tuning examples on Vertex.
+ "contents": [ # Multi-turn contents that represents the Prompt.
+ { # The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message.
+ "parts": [ # Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part.
+ { # A datatype containing media that is part of a multi-part Content message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. For media types that are not text, `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.
+ "audioTranscription": { # The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. # Optional. Audio (input or output) transcription. This is only set when this `Part` contains audio data.
+ "speakerLabel": "A String", # Optional. A label identifying the speaker of this audio segment (e.g. `spk_1`, `spk_2`). Present when `diarization` is set.
+ "text": "A String", # Required. The transcription text of this audio segment.
+ "words": [ # Optional. Detailed word-level transcriptions and timing details. Present when `word_timestamp` is set.
+ { # Information about a single recognized word.
+ "endOffset": "A String", # Optional. End offset in time of the word relative to the start of the audio.
+ "startOffset": "A String", # Optional. Start offset in time of the word relative to the start of the audio.
+ "word": "A String", # Required. Transcript of the word.
+ },
+ ],
},
"codeExecutionResult": { # Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. # Optional. The result of executing the ExecutableCode.
"id": "A String", # Optional. The identifier of the `ExecutableCode` part this result is for. Only populated if the corresponding `ExecutableCode` has an id.
@@ -4949,13 +5657,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -5838,13 +6546,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -7609,6 +8317,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -7834,87 +8719,264 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
- },
- "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
- "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
- "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
- "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
- },
- },
- ],
- "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
- "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
- "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
- },
- },
- },
- "exportLastCheckpointOnly": True or False, # Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.
- "hyperParameters": { # Hyperparameters for SFT. # Optional. Hyperparameters for SFT.
- "adapterSize": "A String", # Optional. Adapter size for tuning.
- "batchSize": "A String", # Optional. Batch size for tuning. This feature is only available for open source models.
- "epochCount": "A String", # Optional. Number of complete passes the model makes over the entire training dataset during training.
- "learningRate": 3.14, # Optional. Learning rate for tuning. Mutually exclusive with `learning_rate_multiplier`. This feature is only available for open source models.
- "learningRateMultiplier": 3.14, # Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with `learning_rate`. This feature is only available for 1P models.
- },
- "trainingDatasetUri": "A String", # Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
- "tuningMode": "A String", # Tuning mode.
- "validationDatasetUri": "A String", # Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
- },
- "tunedModel": { # The Model Registry Model and Online Prediction Endpoint associated with this TuningJob. # Output only. The tuned model resources associated with this TuningJob.
- "checkpoints": [ # Output only. The checkpoints associated with this TunedModel. This field is only populated for tuning jobs that enable intermediate checkpoints.
- { # TunedModelCheckpoint for the Tuned Model of a Tuning Job.
- "checkpointId": "A String", # The ID of the checkpoint.
- "endpoint": "A String", # The Endpoint resource name that the checkpoint is deployed to. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
- "epoch": "A String", # The epoch of the checkpoint.
- "step": "A String", # The step of the checkpoint.
- },
- ],
- "endpoint": "A String", # Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
- "model": "A String", # Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}@{version_id}` When tuning from a base model, the version ID will be 1. For continuous tuning, if the provided tuned_model_display_name is set and different from parent model's display name, the tuned model will have a new parent model with version 1. Otherwise the version id will be incremented by 1 from the last version ID in the parent model. E.g., `projects/{project}/locations/{location}/models/{model}@{last_version_id + 1}`
- },
- "tunedModelDisplayName": "A String", # Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
- "tuningDataStats": { # The tuning data statistic values for TuningJob. # Output only. The tuning data statistics associated with this TuningJob.
- "distillationDataStats": { # Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model. # Output only. Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model.
- "trainingDatasetStats": { # Statistics computed over a tuning dataset. # Output only. Statistics computed for the training dataset.
- "contentsPerExampleDistribution": { # Distribution computed over a tuning dataset. # Output only. Dataset distributions for the number of contents per example.
- "buckets": [ # Output only. Defines the histogram bucket.
- { # Dataset bucket used to create a histogram for the distribution given a population of values.
- "count": "A String", # Output only. Number of values in the bucket.
- "left": 3.14, # Output only. Left bound of the bucket.
- "right": 3.14, # Output only. Right bound of the bucket.
- },
- ],
- "max": 3.14, # Output only. The maximum of the population values.
- "mean": 3.14, # Output only. The arithmetic mean of the values in the population.
- "median": 3.14, # Output only. The median of the values in the population.
- "min": 3.14, # Output only. The minimum of the population values.
- "p5": 3.14, # Output only. The 5th percentile of the values in the population.
- "p95": 3.14, # Output only. The 95th percentile of the values in the population.
- "sum": 3.14, # Output only. Sum of a given population of values.
- },
- "droppedExampleIndices": [ # Output only. A partial sample of the indices (starting from 1) of the dropped examples.
- "A String",
- ],
- "droppedExampleReasons": [ # Output only. For each index in `dropped_example_indices`, the user-facing reason why the example was dropped.
- "A String",
- ],
- "reinforcementTuningUserDatasetExamples": { # Sample reinforcement tuning user data in the training dataset. The contents are truncated for better UI showing. # Output only. Sample user dataset examples in the training dataset uri for Reinforcement Tuning.
- "userDatasetExamples": [ # List of user datasset examples showing to user.
- { # User-facing format for Gemini Reinforcement Tuning examples on Vertex.
- "contents": [ # Multi-turn contents that represents the Prompt.
- { # The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message.
- "parts": [ # Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part.
- { # A datatype containing media that is part of a multi-part Content message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. For media types that are not text, `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.
- "audioTranscription": { # The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. # Optional. Audio (input or output) transcription. This is only set when this `Part` contains audio data.
- "speakerLabel": "A String", # Optional. A label identifying the speaker of this audio segment (e.g. `spk_1`, `spk_2`). Present when `diarization` is set.
- "text": "A String", # Required. The transcription text of this audio segment.
- "words": [ # Optional. Detailed word-level transcriptions and timing details. Present when `word_timestamp` is set.
- { # Information about a single recognized word.
- "endOffset": "A String", # Optional. End offset in time of the word relative to the start of the audio.
- "startOffset": "A String", # Optional. Start offset in time of the word relative to the start of the audio.
- "word": "A String", # Required. Transcript of the word.
- },
- ],
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
+ "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
+ "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
+ "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
+ },
+ },
+ ],
+ "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
+ "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
+ "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
+ },
+ },
+ },
+ "exportLastCheckpointOnly": True or False, # Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.
+ "hyperParameters": { # Hyperparameters for SFT. # Optional. Hyperparameters for SFT.
+ "adapterSize": "A String", # Optional. Adapter size for tuning.
+ "batchSize": "A String", # Optional. Batch size for tuning. This feature is only available for open source models.
+ "epochCount": "A String", # Optional. Number of complete passes the model makes over the entire training dataset during training.
+ "learningRate": 3.14, # Optional. Learning rate for tuning. Mutually exclusive with `learning_rate_multiplier`. This feature is only available for open source models.
+ "learningRateMultiplier": 3.14, # Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with `learning_rate`. This feature is only available for 1P models.
+ },
+ "trainingDatasetUri": "A String", # Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
+ "tuningMode": "A String", # Tuning mode.
+ "validationDatasetUri": "A String", # Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.
+ },
+ "tunedModel": { # The Model Registry Model and Online Prediction Endpoint associated with this TuningJob. # Output only. The tuned model resources associated with this TuningJob.
+ "checkpoints": [ # Output only. The checkpoints associated with this TunedModel. This field is only populated for tuning jobs that enable intermediate checkpoints.
+ { # TunedModelCheckpoint for the Tuned Model of a Tuning Job.
+ "checkpointId": "A String", # The ID of the checkpoint.
+ "endpoint": "A String", # The Endpoint resource name that the checkpoint is deployed to. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "epoch": "A String", # The epoch of the checkpoint.
+ "step": "A String", # The step of the checkpoint.
+ },
+ ],
+ "endpoint": "A String", # Output only. A resource name of an Endpoint. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`.
+ "model": "A String", # Output only. The resource name of the TunedModel. Format: `projects/{project}/locations/{location}/models/{model}@{version_id}` When tuning from a base model, the version ID will be 1. For continuous tuning, if the provided tuned_model_display_name is set and different from parent model's display name, the tuned model will have a new parent model with version 1. Otherwise the version id will be incremented by 1 from the last version ID in the parent model. E.g., `projects/{project}/locations/{location}/models/{model}@{last_version_id + 1}`
+ },
+ "tunedModelDisplayName": "A String", # Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
+ "tuningDataStats": { # The tuning data statistic values for TuningJob. # Output only. The tuning data statistics associated with this TuningJob.
+ "distillationDataStats": { # Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model. # Output only. Statistics for distillation prompt dataset. These statistics do not include the responses sampled from the teacher model.
+ "trainingDatasetStats": { # Statistics computed over a tuning dataset. # Output only. Statistics computed for the training dataset.
+ "contentsPerExampleDistribution": { # Distribution computed over a tuning dataset. # Output only. Dataset distributions for the number of contents per example.
+ "buckets": [ # Output only. Defines the histogram bucket.
+ { # Dataset bucket used to create a histogram for the distribution given a population of values.
+ "count": "A String", # Output only. Number of values in the bucket.
+ "left": 3.14, # Output only. Left bound of the bucket.
+ "right": 3.14, # Output only. Right bound of the bucket.
+ },
+ ],
+ "max": 3.14, # Output only. The maximum of the population values.
+ "mean": 3.14, # Output only. The arithmetic mean of the values in the population.
+ "median": 3.14, # Output only. The median of the values in the population.
+ "min": 3.14, # Output only. The minimum of the population values.
+ "p5": 3.14, # Output only. The 5th percentile of the values in the population.
+ "p95": 3.14, # Output only. The 95th percentile of the values in the population.
+ "sum": 3.14, # Output only. Sum of a given population of values.
+ },
+ "droppedExampleIndices": [ # Output only. A partial sample of the indices (starting from 1) of the dropped examples.
+ "A String",
+ ],
+ "droppedExampleReasons": [ # Output only. For each index in `dropped_example_indices`, the user-facing reason why the example was dropped.
+ "A String",
+ ],
+ "reinforcementTuningUserDatasetExamples": { # Sample reinforcement tuning user data in the training dataset. The contents are truncated for better UI showing. # Output only. Sample user dataset examples in the training dataset uri for Reinforcement Tuning.
+ "userDatasetExamples": [ # List of user datasset examples showing to user.
+ { # User-facing format for Gemini Reinforcement Tuning examples on Vertex.
+ "contents": [ # Multi-turn contents that represents the Prompt.
+ { # The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message.
+ "parts": [ # Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part.
+ { # A datatype containing media that is part of a multi-part Content message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. For media types that are not text, `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.
+ "audioTranscription": { # The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. # Optional. Audio (input or output) transcription. This is only set when this `Part` contains audio data.
+ "speakerLabel": "A String", # Optional. A label identifying the speaker of this audio segment (e.g. `spk_1`, `spk_2`). Present when `diarization` is set.
+ "text": "A String", # Required. The transcription text of this audio segment.
+ "words": [ # Optional. Detailed word-level transcriptions and timing details. Present when `word_timestamp` is set.
+ { # Information about a single recognized word.
+ "endOffset": "A String", # Optional. End offset in time of the word relative to the start of the audio.
+ "startOffset": "A String", # Optional. Start offset in time of the word relative to the start of the audio.
+ "word": "A String", # Required. Transcript of the word.
+ },
+ ],
},
"codeExecutionResult": { # Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. # Optional. The result of executing the ExecutableCode.
"id": "A String", # Optional. The identifier of the `ExecutableCode` part this result is for. Only populated if the corresponding `ExecutableCode` has an id.
@@ -8098,13 +9160,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -8987,13 +10049,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -10756,23 +11818,418 @@ Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
},
},
- "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
+ "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
+ "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
+ "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
+ },
+ },
+ "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
+ "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
+ "rubricContentType": "A String", # The type of rubric content to be generated.
+ "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
+ "A String",
+ ],
},
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
- },
+ "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
+ "systemInstruction": "A String", # Optional. System instructions for the judge model.
+ },
+ "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
+ "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
+ "a_key": "", # Properties of the object.
},
- "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
- "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
- "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
- "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
- },
+ "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
+ "description": "A String", # Optional. The description of the score explaining the directionality etc.
+ "max": 3.14, # Required. The maximum value of the score range (inclusive).
+ "min": 3.14, # Required. The minimum value of the score range (inclusive).
+ "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
},
- "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
- "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
+ },
+ "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
+ "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
+ "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
+ },
+ "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
+ },
+ "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
"autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
"flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
"generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
@@ -10947,47 +12404,6 @@ Method Details
},
"samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
- "rubricContentType": "A String", # The type of rubric content to be generated.
- "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
- "A String",
- ],
- },
- "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
- "systemInstruction": "A String", # Optional. System instructions for the judge model.
- },
- "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
- "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
- "a_key": "", # Properties of the object.
- },
- "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
- "description": "A String", # Optional. The description of the score explaining the directionality etc.
- "max": 3.14, # Required. The maximum value of the score range (inclusive).
- "min": 3.14, # Required. The minimum value of the score range (inclusive).
- "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
- },
- "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
- },
- "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
- "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
- "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
- },
- "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
- },
- "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
@@ -11253,13 +12669,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -12142,13 +13558,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -14162,45 +15578,222 @@ Method Details
"mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
"voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
},
- "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
},
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
},
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
],
- },
- "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
- "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
- "voiceName": "A String", # The name of the prebuilt voice to use.
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
},
- "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
- "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
- "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
},
- "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
},
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
- "A String",
- ],
- "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
- "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
- "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
- "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
- "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
- },
- "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
- "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
- "translationConfig": { # Config for translation features. # Optional. Config for translation.
- "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
- "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
- },
- },
- "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
- },
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
@@ -14427,6 +16020,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
"rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
@@ -14691,13 +16461,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -15580,13 +17350,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -16628,13 +18398,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
diff --git a/docs/dyn/aiplatform_v1beta1.projects.locations.voices.html b/docs/dyn/aiplatform_v1beta1.projects.locations.voices.html
index de0413a4ab..5dee9f4049 100644
--- a/docs/dyn/aiplatform_v1beta1.projects.locations.voices.html
+++ b/docs/dyn/aiplatform_v1beta1.projects.locations.voices.html
@@ -108,13 +108,13 @@ Method Details
The object takes the form of:
{ # Request message for `VoicesService.CreateVoice`.
- "store": True or False, # Optional. Whether the created voice is persisted and managed by Google. * When `true`, Google stores the voice and returns `Voice.id` (for example, `voice_abc123def456`), which can be managed via `GetVoice`, `ListVoices`, and `DeleteVoice` and referenced by ID in synthesis requests. Projects are subject to a maximum active stored voice quota; exceeding the quota returns `RESOURCE_EXHAUSTED`. * When `false` (default), the voice is not stored by Google and `Voice.key` (for example, `voicekey_...`) is returned for client-side storage and synthesis. Optional discovery metadata fields on `voice` are not persisted or returned when `store` is `false`. * Required to be `true` when `voice.type` is `"prompted"` (otherwise fails with `INVALID_ARGUMENT`).
+ "store": True or False, # Optional. Whether the created voice is persisted and managed by Google. * When `true`, Google stores the voice and returns `Voice.id` (for example, `voice_abc123def456`), which can be managed via `GetVoice`, `ListVoices`, and `DeleteVoice` and referenced by ID in synthesis requests. Stored voices expire after 1 year of inactivity; using a stored voice in speech synthesis or as a `base_voice` in `CreateVoice` extends its `expire_time`. Projects are subject to a maximum active stored voice quota; exceeding the quota returns `RESOURCE_EXHAUSTED`. * When `false` (default), the voice is not stored by Google and `Voice.key` (for example, `voicekey_...`) is returned for client-side storage and synthesis. Optional discovery metadata fields on `voice` are not persisted or returned when `store` is `false`. * Required to be `true` when `voice.type` is `"prompted"` (otherwise fails with `INVALID_ARGUMENT`).
"voice": { # A voice resource representing either a custom voice (created via `CreateVoice`) or a prebuilt system voice (returned by `ListVoices`). # Required. The voice to create. `voice.model` is optional; if omitted, the service selects the default voice creation model. Output-only fields on `Voice` (`id`, `key`, `expire_time`, `usage`) are ignored if set.
"accent": "A String", # Optional. Regional accent descriptor (e.g. "American", "British").
"context": "A String", # Optional. Optimal usage context or domain (e.g. "Conversational", "Audiobook", "News").
"description": "A String", # Optional. Descriptive summary of vocal timbre, personality, and tone.
"displayName": "A String", # Optional. User-provided display name for a stored voice (`store = true`), or the catalog name for a prebuilt voice.
- "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
+ "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. For custom stored voices (`store = true`), this expiration time is extended when the voice is used for speech synthesis or as a `base_voice` in `CreateVoice`. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
"gender": "A String", # Optional. Perceived voice gender presentation (e.g. "female", "male", "neutral").
"id": "A String", # Output only. The unique identifier of the voice. * For Google-managed custom voices (`store = true`), this is a generated ID with prefix `voice_` (for example, `voice_abc123def456`). Pass `voices/{id}` as the `name` in `GetVoice` and `DeleteVoice`, and pass `{id}` directly to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis. * For prebuilt catalog voices (`"prebuilt"` returned by `ListVoices`), this is the speaker name (for example, `Puck` or `Charon`). * Unset when `CreateVoice` is called with `store = false`.
"key": "A String", # Output only. The client-managed voice replication key (with prefix `voicekey_`). Returned only by `CreateVoice` when `type` is `"replicated"` and `store` is `false`. Pass this key to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis.
@@ -195,7 +195,7 @@ Method Details
"context": "A String", # Optional. Optimal usage context or domain (e.g. "Conversational", "Audiobook", "News").
"description": "A String", # Optional. Descriptive summary of vocal timbre, personality, and tone.
"displayName": "A String", # Optional. User-provided display name for a stored voice (`store = true`), or the catalog name for a prebuilt voice.
- "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
+ "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. For custom stored voices (`store = true`), this expiration time is extended when the voice is used for speech synthesis or as a `base_voice` in `CreateVoice`. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
"gender": "A String", # Optional. Perceived voice gender presentation (e.g. "female", "male", "neutral").
"id": "A String", # Output only. The unique identifier of the voice. * For Google-managed custom voices (`store = true`), this is a generated ID with prefix `voice_` (for example, `voice_abc123def456`). Pass `voices/{id}` as the `name` in `GetVoice` and `DeleteVoice`, and pass `{id}` directly to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis. * For prebuilt catalog voices (`"prebuilt"` returned by `ListVoices`), this is the speaker name (for example, `Puck` or `Charon`). * Unset when `CreateVoice` is called with `store = false`.
"key": "A String", # Output only. The client-managed voice replication key (with prefix `voicekey_`). Returned only by `CreateVoice` when `type` is `"replicated"` and `store` is `false`. Pass this key to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis.
@@ -300,7 +300,7 @@ Method Details
"context": "A String", # Optional. Optimal usage context or domain (e.g. "Conversational", "Audiobook", "News").
"description": "A String", # Optional. Descriptive summary of vocal timbre, personality, and tone.
"displayName": "A String", # Optional. User-provided display name for a stored voice (`store = true`), or the catalog name for a prebuilt voice.
- "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
+ "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. For custom stored voices (`store = true`), this expiration time is extended when the voice is used for speech synthesis or as a `base_voice` in `CreateVoice`. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
"gender": "A String", # Optional. Perceived voice gender presentation (e.g. "female", "male", "neutral").
"id": "A String", # Output only. The unique identifier of the voice. * For Google-managed custom voices (`store = true`), this is a generated ID with prefix `voice_` (for example, `voice_abc123def456`). Pass `voices/{id}` as the `name` in `GetVoice` and `DeleteVoice`, and pass `{id}` directly to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis. * For prebuilt catalog voices (`"prebuilt"` returned by `ListVoices`), this is the speaker name (for example, `Puck` or `Charon`). * Unset when `CreateVoice` is called with `store = false`.
"key": "A String", # Output only. The client-managed voice replication key (with prefix `voicekey_`). Returned only by `CreateVoice` when `type` is `"replicated"` and `store` is `false`. Pass this key to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis.
@@ -401,7 +401,7 @@ Method Details
"context": "A String", # Optional. Optimal usage context or domain (e.g. "Conversational", "Audiobook", "News").
"description": "A String", # Optional. Descriptive summary of vocal timbre, personality, and tone.
"displayName": "A String", # Optional. User-provided display name for a stored voice (`store = true`), or the catalog name for a prebuilt voice.
- "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
+ "expireTime": "A String", # Output only. The timestamp at which a custom stored voice (`store = true`) or replicated voice key (`store = false`) expires. For custom stored voices (`store = true`), this expiration time is extended when the voice is used for speech synthesis or as a `base_voice` in `CreateVoice`. Unset for prebuilt catalog voices (`"prebuilt"`), which do not expire.
"gender": "A String", # Optional. Perceived voice gender presentation (e.g. "female", "male", "neutral").
"id": "A String", # Output only. The unique identifier of the voice. * For Google-managed custom voices (`store = true`), this is a generated ID with prefix `voice_` (for example, `voice_abc123def456`). Pass `voices/{id}` as the `name` in `GetVoice` and `DeleteVoice`, and pass `{id}` directly to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis. * For prebuilt catalog voices (`"prebuilt"` returned by `ListVoices`), this is the speaker name (for example, `Puck` or `Charon`). * Unset when `CreateVoice` is called with `store = false`.
"key": "A String", # Output only. The client-managed voice replication key (with prefix `voicekey_`). Returned only by `CreateVoice` when `type` is `"replicated"` and `store` is `false`. Pass this key to `SpeechConfig.voice_config.voice` (or `SpeechConfig.voice`) during speech synthesis.
diff --git a/docs/dyn/aiplatform_v1beta1.publishers.models.html b/docs/dyn/aiplatform_v1beta1.publishers.models.html
index 4dffefda03..4b46b80748 100644
--- a/docs/dyn/aiplatform_v1beta1.publishers.models.html
+++ b/docs/dyn/aiplatform_v1beta1.publishers.models.html
@@ -626,13 +626,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -1333,13 +1333,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -3622,13 +3622,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
diff --git a/docs/dyn/aiplatform_v1beta1.v1beta1.html b/docs/dyn/aiplatform_v1beta1.v1beta1.html
index d8241afb26..280508c362 100644
--- a/docs/dyn/aiplatform_v1beta1.v1beta1.html
+++ b/docs/dyn/aiplatform_v1beta1.v1beta1.html
@@ -492,6 +492,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -717,87 +894,265 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
- },
- "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
- "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
- "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
- "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
- },
- },
- ],
- "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
- "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
- "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
- },
- },
-}
-
- x__xgafv: string, V1 error format.
- Allowed values
- 1 - v1 error format
- 2 - v2 error format
-
-Returns:
- An object of the form:
-
- { # This resource represents a long-running operation that is the result of a network API call.
- "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available.
- "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation.
- "code": 42, # The status code, which should be an enum value of google.rpc.Code.
- "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use.
- {
- "a_key": "", # Properties of the object. Contains field @type with type URL.
- },
- ],
- "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.
- },
- "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any.
- "a_key": "", # Properties of the object. Contains field @type with type URL.
- },
- "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`.
- "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`.
- "a_key": "", # Properties of the object. Contains field @type with type URL.
- },
-}
-evaluateInstances(body=None, x__xgafv=None)
- Evaluates instances based on a given metric.
-
-Args:
- body: object, The request body.
- The object takes the form of:
-
-{ # Request message for EvaluationService.EvaluateInstances.
- "autoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Autorater config used for evaluation. Not applicable for predefined metrics (PredefinedMetricSpec); the server uses its own model configuration for predefined metrics and this field is ignored.
- "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
- "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
- "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
- "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
- "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
- "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
- "A String",
- ],
- "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
- "A String",
- ],
- "diarization": True or False, # Optional. Configures speaker diarization.
- "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
- },
- "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
- "A String",
- ],
- "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
- "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
- "A String",
- ],
- },
- "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
- "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
- },
- "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
- "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
+ "rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
+ "splitSummaries": True or False, # Optional. Whether to split summaries while using rougeLsum.
+ "useStemmer": True or False, # Optional. Whether to use stemmer to compute rouge score.
+ },
+ },
+ ],
+ "outputConfig": { # Config for evaluation output. # Required. Config for evaluation output.
+ "gcsDestination": { # The Google Cloud Storage location where the output is to be written to. # Cloud storage destination for evaluation output.
+ "outputUriPrefix": "A String", # Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.
+ },
+ },
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # This resource represents a long-running operation that is the result of a network API call.
+ "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available.
+ "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation.
+ "code": 42, # The status code, which should be an enum value of google.rpc.Code.
+ "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use.
+ {
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ ],
+ "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.
+ },
+ "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`.
+ "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+}
+evaluateInstances(body=None, x__xgafv=None)
+ Evaluates instances based on a given metric.
+
+Args:
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request message for EvaluationService.EvaluateInstances.
+ "allowCrossRegionModel": True or False, # Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the request location is provided elsewhere in the request, this flag must be set to true or the request will fail.
+ "autoraterConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Optional. Autorater config used for evaluation. Not applicable for predefined metrics (PredefinedMetricSpec); the server uses its own model configuration for predefined metrics and this field is ignored.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
"frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
"imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
"aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
@@ -1127,13 +1482,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -1482,13 +1837,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -1928,13 +2283,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -2176,13 +2531,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -2534,13 +2889,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -2782,13 +3137,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -3766,15 +4121,410 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
+ },
+ "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
+ "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
+ "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
+ "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
+ },
+ },
+ "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
+ "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
+ "rubricContentType": "A String", # The type of rubric content to be generated.
+ "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
+ "A String",
+ ],
},
- "resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
- "customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
- "codeExecutionRegion": "A String", # Optional. The region to use for code execution. If set, the Code Execution Sandbox will be invoked in the specified region regardless of the request's originating region. Must be a region where the Code Execution Sandbox is available. For the current list of [supported regions](https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/locations). If unset, the request's originating region is used.
- "parsingFunction": "A String", # Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation
- },
+ "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
+ "systemInstruction": "A String", # Optional. System instructions for the judge model.
+ },
+ "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
+ "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
+ "a_key": "", # Properties of the object.
},
- "rubricGenerationSpec": { # Specification for how rubrics should be generated. # Dynamically generate rubrics using this specification.
- "modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
+ "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
+ "description": "A String", # Optional. The description of the score explaining the directionality etc.
+ "max": 3.14, # Required. The maximum value of the score range (inclusive).
+ "min": 3.14, # Required. The minimum value of the score range (inclusive).
+ "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
+ },
+ "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
+ },
+ "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
+ "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
+ "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
+ },
+ "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
+ "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
+ "returnRawOutput": True or False, # Optional. Whether to return raw output.
+ },
+ "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
+ "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
+ },
+ "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
"autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
"flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
"generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
@@ -3949,47 +4699,6 @@ Method Details
},
"samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
},
- "promptTemplate": "A String", # Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements.
- "rubricContentType": "A String", # The type of rubric content to be generated.
- "rubricTypeOntology": [ # Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology.
- "A String",
- ],
- },
- "rubricGroupKey": "A String", # Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance.
- "systemInstruction": "A String", # Optional. System instructions for the judge model.
- },
- "metadata": { # Metadata about the metric, used for visualization and organization. # Optional. Metadata about the metric, used for visualization and organization.
- "otherMetadata": { # Optional. Flexible metadata for user-defined attributes.
- "a_key": "", # Properties of the object.
- },
- "scoreRange": { # The range of possible scores for this metric, used for plotting. # Optional. The range of possible scores for this metric, used for plotting.
- "description": "A String", # Optional. The description of the score explaining the directionality etc.
- "max": 3.14, # Required. The maximum value of the score range (inclusive).
- "min": 3.14, # Required. The minimum value of the score range (inclusive).
- "step": 3.14, # Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous.
- },
- "title": "A String", # Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name.
- },
- "pairwiseMetricSpec": { # Spec for pairwise metric. # Spec for pairwise metric.
- "baselineResponseFieldName": "A String", # Optional. The field name of the baseline response.
- "candidateResponseFieldName": "A String", # Optional. The field name of the candidate response.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. When this config is set, the default output is replaced with the raw output string. If a custom format is chosen, the `pairwise_choice` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pairwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pairwise metric.
- },
- "pointwiseMetricSpec": { # Spec for pointwise metric. # Spec for pointwise metric.
- "customOutputFormatConfig": { # Spec for custom output format configuration. # Optional. CustomOutputFormatConfig allows customization of metric output. By default, metrics return a score and explanation. When this config is set, the default output is replaced with either: - The raw output string. - A parsed output based on a user-defined schema. If a custom format is chosen, the `score` and `explanation` fields in the corresponding metric result will be empty.
- "returnRawOutput": True or False, # Optional. Whether to return raw output.
- },
- "metricPromptTemplate": "A String", # Required. Metric prompt template for pointwise metric.
- "systemInstruction": "A String", # Optional. System instructions for pointwise metric.
- },
- "predefinedMetricSpec": { # The spec for a pre-defined metric. # The spec for a pre-defined metric.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
},
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
@@ -4198,14 +4907,191 @@ Method Details
"targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
},
},
- "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
- },
- "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
- "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
- "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
- "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
- "a_key": "", # Properties of the object.
- },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ "metricPromptTemplate": "A String", # Required. Template for the prompt sent to the judge model.
+ "predefinedRubricGenerationSpec": { # The spec for a pre-defined metric. # Dynamically generate rubrics using a predefined spec.
+ "metricSpecName": "A String", # Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1".
+ "metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
+ "a_key": "", # Properties of the object.
+ },
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"resultParserConfig": { # Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. # Optional. The parser config for the metric result.
"customCodeParserConfig": { # Configuration for parsing the LLM response using custom code. # Optional. Use custom code to parse the LLM response.
@@ -4431,6 +5317,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rougeSpec": { # Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. # Spec for rouge metric.
"rougeType": "A String", # Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
@@ -5362,13 +6425,13 @@ Method Details
{ # Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).
"codeExecution": { # Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. # Optional. CodeExecution tool type. Enables the model to execute code as part of generation.
},
- "computerUse": { # Tool to support computer use. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
- "disabledSafetyPolicies": [ # Optional. Disabled safety policies for computer use.
+ "computerUse": { # A tool that enables the model to interact directly with a computer environment. # Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.
+ "disabledSafetyPolicies": [ # Optional. A list of safety policies to disable for the computer use tool.
"A String",
],
- "enablePromptInjectionDetection": True or False, # Optional. Enables the prompt injection detection check on computer-use request.
- "environment": "A String", # Required. The environment being operated.
- "excludedPredefinedFunctions": [ # Optional. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions.
+ "enablePromptInjectionDetection": True or False, # Optional. Whether to enable the prompt injection detection check on the computer use request.
+ "environment": "A String", # Required. The target environment where the computer use tool operates.
+ "excludedPredefinedFunctions": [ # Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions.
"A String",
],
},
@@ -5706,6 +6769,183 @@ Method Details
"metricSpecParameters": { # Optional. The parameters needed to run the pre-defined metric.
"a_key": "", # Properties of the object.
},
+ "stepAutoraterConfigs": { # Optional. Per-step autorater overrides for this predefined metric. Keys are evaluation stage names (snake_case). Valid keys: "intent_extraction", "rubric_generation", "rubric_validation". Unknown keys return INVALID_ARGUMENT.
+ "a_key": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset.
+ "autoraterModel": "A String", # Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
+ "flipEnabled": True or False, # Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias.
+ "generationConfig": { # Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. # Optional. Configuration options for model generation and outputs.
+ "audioTimestamp": True or False, # Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response.
+ "audioTranscriptionConfig": { # Configuration for speech recognition (transcription). # Optional. Configuration for audio transcription (speech recognition).
+ "adaptationPhrases": [ # Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards.
+ "A String",
+ ],
+ "customVocabulary": [ # Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms.
+ "A String",
+ ],
+ "diarization": True or False, # Optional. Configures speaker diarization.
+ "languageAuto": { # Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. # Optional. Deprecated: Use top-level `language_codes` instead. The model will detect the language automatically.
+ },
+ "languageCodes": [ # Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection.
+ "A String",
+ ],
+ "languageHints": { # Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. # Optional. Deprecated: Use top-level `language_codes` instead. Specifies one or more languages in the audio.
+ "languageCodes": [ # Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified.
+ "A String",
+ ],
+ },
+ "mode": "A String", # Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.
+ "wordTimestamp": True or False, # Optional. Configures word-level timestamp generation.
+ },
+ "candidateCount": 42, # Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one.
+ "enableAffectiveDialog": True or False, # Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response.
+ "frequencyPenalty": 3.14, # Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0].
+ "imageConfig": { # Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. # Optional. Config for image generation features. Deprecated: Use `response_format.image` instead.
+ "aspectRatio": "A String", # Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9"
+ "imageOutputOptions": { # The image output format for generated images. # Optional. The image output format for generated images.
+ "compressionQuality": 42, # Optional. The compression quality of the output image.
+ "mimeType": "A String", # Optional. The image format that the output should be saved as.
+ },
+ "imageSize": "A String", # Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`.
+ "personGeneration": "A String", # Optional. Controls whether the model can generate people.
+ "prominentPeople": "A String", # Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people.
+ },
+ "logprobs": 42, # Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response.
+ "maxOutputTokens": 42, # Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses.
+ "mediaResolution": "A String", # Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model.
+ "modelConfig": { # Config for model selection. # Optional. Config for model selection.
+ "featureSelectionPreference": "A String", # Required. Feature selection preference.
+ },
+ "presencePenalty": 3.14, # Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0].
+ "responseFormat": [ # Optional. New response format field for the model to configure output formatting and delivery.
+ { # Configuration for the model to configure output formatting and delivery.
+ "audio": { # Configuration for audio-specific output formatting. # Audio output format.
+ "bitRate": 42, # Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus).
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "mimeType": "A String", # Optional. The MIME type of the audio output.
+ "sampleRate": 42, # Optional. Sample rate for the generated audio in Hertz.
+ },
+ "image": { # Configuration for image-specific output formatting. # Image output format.
+ "aspectRatio": "A String", # Optional. The aspect ratio for the image output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "imageSize": "A String", # Optional. The size of the image output.
+ "mimeType": "A String", # Optional. The MIME type of the image output.
+ },
+ "text": { # Configuration for text-specific output formatting. # Text output format.
+ "mimeType": "A String", # Optional. The IANA standard MIME type of the response.
+ "schema": "", # Optional. The JSON schema that the output should conform to. Only applicable when mime_type is APPLICATION_JSON.
+ },
+ "video": { # Configuration for video-specific output formatting. # Video output format.
+ "aspectRatio": "A String", # The aspect ratio for the video output.
+ "delivery": "A String", # Optional. Delivery mode for the generated content.
+ "duration": "A String", # Optional. The duration for the video output.
+ "gcsUri": "A String", # Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI.
+ "resolution": "A String", # Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k".
+ },
+ },
+ ],
+ "responseJsonSchema": "", # Optional. When this field is set, response_schema must be omitted and response_mime_type must be set to `application/json`. Deprecated: Use `response_format` instead.
+ "responseLogprobs": True or False, # Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging.
+ "responseMimeType": "A String", # Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead.
+ "responseModalities": [ # Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image.
+ "A String",
+ ],
+ "responseSchema": { # Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). # Optional. Lets you to specify a schema for the model's response, ensuring that the output conforms to a particular structure. This is useful for generating structured data such as JSON. The schema is a subset of the [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema) object. When this field is set, you must also set the `response_mime_type` to `application/json`. Deprecated: Use `response_format` instead.
+ "additionalProperties": "", # Optional. If `type` is `OBJECT`, specifies how to handle properties not defined in `properties`. If it is a boolean `false`, no additional properties are allowed. If it is a schema, additional properties are allowed if they conform to the schema.
+ "anyOf": [ # Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`.
+ # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ ],
+ "default": "", # Optional. Default value to use if the field is not specified.
+ "defs": { # Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "description": "A String", # Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt.
+ "enum": [ # Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}`
+ "A String",
+ ],
+ "example": "", # Optional. Example of an instance of this schema.
+ "format": "A String", # Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type.
+ "items": # Object with schema name: GoogleCloudAiplatformV1beta1Schema # Optional. If type is `ARRAY`, `items` specifies the schema of elements in the array.
+ "maxItems": "A String", # Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array.
+ "maxLength": "A String", # Optional. If type is `STRING`, `max_length` specifies the maximum length of the string.
+ "maxProperties": "A String", # Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided.
+ "maximum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value.
+ "minItems": "A String", # Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array.
+ "minLength": "A String", # Optional. If type is `STRING`, `min_length` specifies the minimum length of the string.
+ "minProperties": "A String", # Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided.
+ "minimum": 3.14, # Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value.
+ "nullable": True or False, # Optional. Indicates if the value of this field can be null.
+ "pattern": "A String", # Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match.
+ "properties": { # Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object.
+ "a_key": # Object with schema name: GoogleCloudAiplatformV1beta1Schema
+ },
+ "propertyOrdering": [ # Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties.
+ "A String",
+ ],
+ "ref": "A String", # Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
+ "required": [ # Optional. If type is `OBJECT`, `required` lists the names of properties that must be present.
+ "A String",
+ ],
+ "title": "A String", # Optional. Title for the schema.
+ "type": "A String", # Optional. Data type of the schema field.
+ },
+ "routingConfig": { # The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. # Optional. Routing configuration.
+ "autoMode": { # The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. # In this mode, the model is selected automatically based on the content of the request.
+ "modelRoutingPreference": "A String", # The model routing preference.
+ },
+ "manualMode": { # The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. # In this mode, the model is specified manually.
+ "modelName": "A String", # The name of the model to use. Only public LLM models are accepted.
+ },
+ },
+ "seed": 42, # Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results.
+ "speechConfig": { # Configuration for speech generation. # Optional. The speech generation config.
+ "languageCode": "A String", # Optional. The language code (ISO 639-1) for the speech synthesis.
+ "multiSpeakerVoiceConfig": { # Configuration for a multi-speaker text-to-speech request. # The configuration for a multi-speaker text-to-speech request. This field is mutually exclusive with `voice_config`.
+ "speakerVoiceConfigs": [ # Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided.
+ { # Configuration for a single speaker in a multi-speaker setup.
+ "speaker": "A String", # Required. The name of the speaker. This should be the same as the speaker name used in the prompt.
+ "voiceConfig": { # Configuration for a voice. # Required. The configuration for the voice of this speaker.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ ],
+ },
+ "voiceConfig": { # Configuration for a voice. # The configuration for the voice to use.
+ "prebuiltVoiceConfig": { # Configuration for a prebuilt voice. # The configuration for a prebuilt voice.
+ "voiceName": "A String", # The name of the prebuilt voice to use.
+ },
+ "replicatedVoiceConfig": { # The configuration for the replicated voice to use. # Optional. The configuration for a replicated voice. This enables users to replicate a voice from an audio sample.
+ "mimeType": "A String", # Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set.
+ "voiceSampleAudio": "A String", # Optional. The sample of the custom voice.
+ },
+ "voice": "A String", # Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`).
+ },
+ },
+ "stopSequences": [ # Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker.
+ "A String",
+ ],
+ "temperature": 3.14, # Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0].
+ "thinkingConfig": { # Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. # Optional. Configuration for thinking features. An error will be returned if this field is set for models that don't support thinking.
+ "includeThoughts": True or False, # Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available.
+ "thinkingBudget": 42, # Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency.
+ "thinkingLevel": "A String", # Optional. The number of thoughts tokens that the model should generate.
+ },
+ "topK": 3.14, # Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words.
+ "topP": 3.14, # Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both.
+ "translationConfig": { # Config for translation features. # Optional. Config for translation.
+ "echoTargetLanguage": True or False, # Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language.
+ "targetLanguageCode": "A String", # Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr").
+ },
+ },
+ "samplingCount": 42, # Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32.
+ },
+ },
},
"rubricGenerationSpec": { # Specification for how rubrics should be generated. # Optional. Specification for how the rubrics should be generated.
"modelConfig": { # The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. # Configuration for the model used in rubric generation. Configs including sampling count and base model can be specified here. Flipping is not supported for rubric generation.
diff --git a/docs/dyn/alloydb_v1.projects.locations.clusters.instances.html b/docs/dyn/alloydb_v1.projects.locations.clusters.instances.html
index 41e0361bb3..863167e8bd 100644
--- a/docs/dyn/alloydb_v1.projects.locations.clusters.instances.html
+++ b/docs/dyn/alloydb_v1.projects.locations.clusters.instances.html
@@ -250,7 +250,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -265,6 +265,27 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
+ "coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
+ "cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
+ "utilizationTarget": 3.14, # Target CPU utilization as a float between 0 and 1.
+ },
+ "enabled": True or False, # If true, autoscaling is enabled for the instance. If not set, the default value is false.
+ "maxNodeCount": "A String", # Maximum number of nodes for the autoscaler.
+ },
+ "schedules": [ # Optional list of schedules for the MIG autoscaler. If not set, no schedules are created.
+ { # A schedule for the autoscaler.
+ "cronExpression": "A String", # Cron expression for the triggering the schedule. See https://cloud.google.com/compute/docs/autoscaler/scaling-schedules#cron_expressions for the syntax.
+ "description": "A String", # Description of the schedule.
+ "disabled": True or False, # If true, the schedule is disabled.
+ "durationSec": "A String", # Duration of the schedule.
+ "minNodeCount": "A String", # Minimum number of nodes in while the schedule is active.
+ "name": "A String", # Name of the schedule.
+ "timeZone": "A String", # The location-based IANA time zone for interpreting the schedule's start time. If no time zone is provided, UTC is used by default.
+ },
+ ],
+ },
"nodeCount": 42, # Read capacity, i.e. number of nodes in a read pool instance.
},
"reconciling": True or False, # Output only. Reconciling (https://google.aip.dev/128#reconciliation). Set to true if the current state of Instance does not match the user's intended state, and the service is actively updating the resource to reconcile them. This can happen due to user-triggered updates or system actions like failover or maintenance.
@@ -447,7 +468,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -462,6 +483,27 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
+ "coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
+ "cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
+ "utilizationTarget": 3.14, # Target CPU utilization as a float between 0 and 1.
+ },
+ "enabled": True or False, # If true, autoscaling is enabled for the instance. If not set, the default value is false.
+ "maxNodeCount": "A String", # Maximum number of nodes for the autoscaler.
+ },
+ "schedules": [ # Optional list of schedules for the MIG autoscaler. If not set, no schedules are created.
+ { # A schedule for the autoscaler.
+ "cronExpression": "A String", # Cron expression for the triggering the schedule. See https://cloud.google.com/compute/docs/autoscaler/scaling-schedules#cron_expressions for the syntax.
+ "description": "A String", # Description of the schedule.
+ "disabled": True or False, # If true, the schedule is disabled.
+ "durationSec": "A String", # Duration of the schedule.
+ "minNodeCount": "A String", # Minimum number of nodes in while the schedule is active.
+ "name": "A String", # Name of the schedule.
+ "timeZone": "A String", # The location-based IANA time zone for interpreting the schedule's start time. If no time zone is provided, UTC is used by default.
+ },
+ ],
+ },
"nodeCount": 42, # Read capacity, i.e. number of nodes in a read pool instance.
},
"reconciling": True or False, # Output only. Reconciling (https://google.aip.dev/128#reconciliation). Set to true if the current state of Instance does not match the user's intended state, and the service is actively updating the resource to reconcile them. This can happen due to user-triggered updates or system actions like failover or maintenance.
@@ -735,7 +777,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -750,6 +792,27 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
+ "coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
+ "cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
+ "utilizationTarget": 3.14, # Target CPU utilization as a float between 0 and 1.
+ },
+ "enabled": True or False, # If true, autoscaling is enabled for the instance. If not set, the default value is false.
+ "maxNodeCount": "A String", # Maximum number of nodes for the autoscaler.
+ },
+ "schedules": [ # Optional list of schedules for the MIG autoscaler. If not set, no schedules are created.
+ { # A schedule for the autoscaler.
+ "cronExpression": "A String", # Cron expression for the triggering the schedule. See https://cloud.google.com/compute/docs/autoscaler/scaling-schedules#cron_expressions for the syntax.
+ "description": "A String", # Description of the schedule.
+ "disabled": True or False, # If true, the schedule is disabled.
+ "durationSec": "A String", # Duration of the schedule.
+ "minNodeCount": "A String", # Minimum number of nodes in while the schedule is active.
+ "name": "A String", # Name of the schedule.
+ "timeZone": "A String", # The location-based IANA time zone for interpreting the schedule's start time. If no time zone is provided, UTC is used by default.
+ },
+ ],
+ },
"nodeCount": 42, # Read capacity, i.e. number of nodes in a read pool instance.
},
"reconciling": True or False, # Output only. Reconciling (https://google.aip.dev/128#reconciliation). Set to true if the current state of Instance does not match the user's intended state, and the service is actively updating the resource to reconcile them. This can happen due to user-triggered updates or system actions like failover or maintenance.
@@ -979,7 +1042,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -994,6 +1057,27 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
+ "coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
+ "cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
+ "utilizationTarget": 3.14, # Target CPU utilization as a float between 0 and 1.
+ },
+ "enabled": True or False, # If true, autoscaling is enabled for the instance. If not set, the default value is false.
+ "maxNodeCount": "A String", # Maximum number of nodes for the autoscaler.
+ },
+ "schedules": [ # Optional list of schedules for the MIG autoscaler. If not set, no schedules are created.
+ { # A schedule for the autoscaler.
+ "cronExpression": "A String", # Cron expression for the triggering the schedule. See https://cloud.google.com/compute/docs/autoscaler/scaling-schedules#cron_expressions for the syntax.
+ "description": "A String", # Description of the schedule.
+ "disabled": True or False, # If true, the schedule is disabled.
+ "durationSec": "A String", # Duration of the schedule.
+ "minNodeCount": "A String", # Minimum number of nodes in while the schedule is active.
+ "name": "A String", # Name of the schedule.
+ "timeZone": "A String", # The location-based IANA time zone for interpreting the schedule's start time. If no time zone is provided, UTC is used by default.
+ },
+ ],
+ },
"nodeCount": 42, # Read capacity, i.e. number of nodes in a read pool instance.
},
"reconciling": True or False, # Output only. Reconciling (https://google.aip.dev/128#reconciliation). Set to true if the current state of Instance does not match the user's intended state, and the service is actively updating the resource to reconcile them. This can happen due to user-triggered updates or system actions like failover or maintenance.
@@ -1165,7 +1249,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -1180,6 +1264,27 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
+ "coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
+ "cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
+ "utilizationTarget": 3.14, # Target CPU utilization as a float between 0 and 1.
+ },
+ "enabled": True or False, # If true, autoscaling is enabled for the instance. If not set, the default value is false.
+ "maxNodeCount": "A String", # Maximum number of nodes for the autoscaler.
+ },
+ "schedules": [ # Optional list of schedules for the MIG autoscaler. If not set, no schedules are created.
+ { # A schedule for the autoscaler.
+ "cronExpression": "A String", # Cron expression for the triggering the schedule. See https://cloud.google.com/compute/docs/autoscaler/scaling-schedules#cron_expressions for the syntax.
+ "description": "A String", # Description of the schedule.
+ "disabled": True or False, # If true, the schedule is disabled.
+ "durationSec": "A String", # Duration of the schedule.
+ "minNodeCount": "A String", # Minimum number of nodes in while the schedule is active.
+ "name": "A String", # Name of the schedule.
+ "timeZone": "A String", # The location-based IANA time zone for interpreting the schedule's start time. If no time zone is provided, UTC is used by default.
+ },
+ ],
+ },
"nodeCount": 42, # Read capacity, i.e. number of nodes in a read pool instance.
},
"reconciling": True or False, # Output only. Reconciling (https://google.aip.dev/128#reconciliation). Set to true if the current state of Instance does not match the user's intended state, and the service is actively updating the resource to reconcile them. This can happen due to user-triggered updates or system actions like failover or maintenance.
diff --git a/docs/dyn/alloydb_v1alpha.projects.locations.backups.html b/docs/dyn/alloydb_v1alpha.projects.locations.backups.html
index e375ecff39..fccb5abf12 100644
--- a/docs/dyn/alloydb_v1alpha.projects.locations.backups.html
+++ b/docs/dyn/alloydb_v1alpha.projects.locations.backups.html
@@ -123,7 +123,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -139,6 +139,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
@@ -258,7 +261,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -274,6 +277,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
@@ -330,7 +336,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -346,6 +352,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
@@ -406,7 +415,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -422,6 +431,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
diff --git a/docs/dyn/alloydb_v1alpha.projects.locations.clusters.html b/docs/dyn/alloydb_v1alpha.projects.locations.clusters.html
index 349b628c6a..a1ae57f8d5 100644
--- a/docs/dyn/alloydb_v1alpha.projects.locations.clusters.html
+++ b/docs/dyn/alloydb_v1alpha.projects.locations.clusters.html
@@ -244,7 +244,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -492,7 +492,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -848,7 +848,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1131,7 +1131,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1368,7 +1368,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1674,7 +1674,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1933,7 +1933,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
diff --git a/docs/dyn/alloydb_v1alpha.projects.locations.clusters.instances.html b/docs/dyn/alloydb_v1alpha.projects.locations.clusters.instances.html
index 53d9cbe62e..a610eaa881 100644
--- a/docs/dyn/alloydb_v1alpha.projects.locations.clusters.instances.html
+++ b/docs/dyn/alloydb_v1alpha.projects.locations.clusters.instances.html
@@ -262,7 +262,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -277,7 +277,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -496,7 +496,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -511,7 +511,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -821,7 +821,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -836,7 +836,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -1107,7 +1107,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -1122,7 +1122,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -1330,7 +1330,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -1345,7 +1345,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
diff --git a/docs/dyn/alloydb_v1beta.projects.locations.backups.html b/docs/dyn/alloydb_v1beta.projects.locations.backups.html
index 39bb20731a..24d879a011 100644
--- a/docs/dyn/alloydb_v1beta.projects.locations.backups.html
+++ b/docs/dyn/alloydb_v1beta.projects.locations.backups.html
@@ -123,7 +123,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -139,6 +139,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
@@ -257,7 +260,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -273,6 +276,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
@@ -328,7 +334,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -344,6 +350,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
@@ -403,7 +412,7 @@ Method Details
"deleteTime": "A String", # Output only. Delete time stamp
"description": "A String", # User-provided description of the backup.
"displayName": "A String", # User-settable and human-readable display name for the Backup.
- "edition": "A String", # Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
+ "edition": "A String", # Output only. Deprecated: Use `express_config` instead. Output only. The edition of the cluster this backup was created from. Any restored cluster created from this backup will have the same edition.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the backup with a customer-managed encryption key (CMEK). When this field is not specified, the backup will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -419,6 +428,9 @@ Method Details
"totalRetentionCount": 42, # Output only. The length of the quantity-based queue, specified by the backup's retention policy.
},
"expiryTime": "A String", # Output only. The time at which after the backup is eligible to be garbage collected. It is the duration specified by the backup's retention policy, added to the backup's create_time.
+ "expressConfig": { # Configuration that allows the user to create an AlloyDB Express cluster. # Output only. The Express configuration of the cluster this backup was created from. Any restored cluster created from this backup will have the same Express configuration.
+ "enabled": True or False, # Optional. Whether Express configuration is enabled for the cluster.
+ },
"labels": { # Labels as key value pairs
"a_key": "A String",
},
diff --git a/docs/dyn/alloydb_v1beta.projects.locations.clusters.html b/docs/dyn/alloydb_v1beta.projects.locations.clusters.html
index 28f11824db..f146fc0c48 100644
--- a/docs/dyn/alloydb_v1beta.projects.locations.clusters.html
+++ b/docs/dyn/alloydb_v1beta.projects.locations.clusters.html
@@ -244,7 +244,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -491,7 +491,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -846,7 +846,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1128,7 +1128,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1364,7 +1364,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1669,7 +1669,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
@@ -1927,7 +1927,7 @@ Method Details
},
"deleteTime": "A String", # Output only. Delete time stamp
"displayName": "A String", # User-settable and human-readable display name for the Cluster.
- "edition": "A String", # Optional. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
+ "edition": "A String", # Optional. Deprecated: Use `express_config` instead. Edition of the cluster. If left unspecified, the cluster behaves as `EDITION_ALLOYDB`.
"encryptionConfig": { # EncryptionConfig describes the encryption config of a cluster or a backup that is encrypted with a CMEK (customer-managed encryption key). # Optional. The encryption config can be specified to encrypt the data disks and other persistent data resources of a cluster with a customer-managed encryption key (CMEK). When this field is not specified, the cluster will then use default encryption scheme to protect the user data.
"kmsKeyName": "A String", # The fully-qualified resource name of the KMS key. Each Cloud KMS key is regionalized and has the following format: projects/[PROJECT]/locations/[REGION]/keyRings/[RING]/cryptoKeys/[KEY_NAME]
},
diff --git a/docs/dyn/alloydb_v1beta.projects.locations.clusters.instances.html b/docs/dyn/alloydb_v1beta.projects.locations.clusters.instances.html
index c5267b1768..51080b867a 100644
--- a/docs/dyn/alloydb_v1beta.projects.locations.clusters.instances.html
+++ b/docs/dyn/alloydb_v1beta.projects.locations.clusters.instances.html
@@ -262,7 +262,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -277,7 +277,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -495,7 +495,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -510,7 +510,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -819,7 +819,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -834,7 +834,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -1104,7 +1104,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -1119,7 +1119,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
@@ -1326,7 +1326,7 @@ Method Details
],
"serviceConnectionPolicy": "A String", # Output only. The PSC service connection policy name. The format is "projects//regions//serviceConnectionPolicies/"
},
- "publicIpAddress": "A String", # Output only. The public IP addresses for the Instance. This is available ONLY when enable_public_ip is set. This is the connection endpoint for an end-user application.
+ "publicIpAddress": "A String", # Output only. The public IP address for the Instance. This is available ONLY when `enable_public_ip` is set or when you have upgraded from an Express Instance that has public DNS enabled. This field is not available for Express Instances. This is the connection endpoint for an end-user application.
"queryInsightsConfig": { # QueryInsights Instance specific configuration. # Configuration for query insights.
"queryPlansPerMinute": 42, # Number of query execution plans captured by Insights per minute for all queries combined. The default value is 5. Any integer between 0 and 20 is considered valid.
"queryStringLength": 42, # Query string length. The default value is 1024. Any integer between 256 and 4500 is considered valid.
@@ -1341,7 +1341,7 @@ Method Details
"recordClientAddress": True or False, # Output only. Whether to record client address.
},
"readPoolConfig": { # Configuration for a read pool instance. # Read pool instance configuration. This is required if the value of instanceType is READ_POOL.
- "autoScalingConfig": { # Configuration for autoscaling. # Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
+ "autoScalingConfig": { # Configuration for autoscaling. # Optional. Autoscaling configuration for the read pool instance. If not set, the read pool instance will not be autoscaled.
"policy": { # Policy for the autoscaler. # Policy for the MIG autoscaler.
"coolDownPeriodSec": "A String", # The period of time in seconds after a new node is created before the autoscaler will incorporate its resource usage (e.g. CPU utilization) into the autoscaling recommendation algorithm.
"cpuUtilization": { # CPU utilization policy for the autoscaler. # CPU utilization policy for the autoscaler.
diff --git a/docs/dyn/analyticsdata_v1beta.properties.html b/docs/dyn/analyticsdata_v1beta.properties.html
index 5229fe05cf..d0895b3dfb 100644
--- a/docs/dyn/analyticsdata_v1beta.properties.html
+++ b/docs/dyn/analyticsdata_v1beta.properties.html
@@ -157,7 +157,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -214,7 +214,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -280,7 +280,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -541,7 +541,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -598,7 +598,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -668,7 +668,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -908,7 +908,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -973,7 +973,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1169,7 +1169,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1226,7 +1226,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1292,7 +1292,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1520,7 +1520,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1589,7 +1589,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1814,7 +1814,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1871,7 +1871,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
@@ -1941,7 +1941,7 @@ Method Details
"int64Value": "A String", # Integer value
},
},
- "emptyFilter": { # Filter for empty values. # A filter for empty values such as "(not set)" and "" values.
+ "emptyFilter": { # Filter for empty values. Matches dimension values that are `""` or `(not set)` values. Use this filter to match rows with missing or unpopulated dimension values, or combine it with `not_expression` to exclude them from a report without filtering for both `""` and `(not set)` separately. # A filter for empty values such as `(not set)` and `""` values.
},
"fieldName": "A String", # The dimension name or metric name. In most methods, dimensions & metrics can be used for the first time in this field. However in a RunPivotReportRequest, this field must be additionally specified by name in the RunPivotReportRequest's dimensions or metrics.
"inListFilter": { # The result needs to be in a list of string values. # A filter for in list values.
diff --git a/docs/dyn/androidmanagement_v1.enterprises.policies.html b/docs/dyn/androidmanagement_v1.enterprises.policies.html
index f5a315d039..fa7696d05a 100644
--- a/docs/dyn/androidmanagement_v1.enterprises.policies.html
+++ b/docs/dyn/androidmanagement_v1.enterprises.policies.html
@@ -407,7 +407,7 @@ Method Details
"A String",
],
"kioskCustomLauncherEnabled": True or False, # Whether the kiosk custom launcher is enabled. This replaces the home screen with a launcher that locks down the device to the apps installed via the applications setting. Apps appear on a single page in alphabetical order. Use kioskCustomization to further configure the kiosk device behavior.
- "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK.
+ "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role.
"deviceSettings": "A String", # Optional. Specifies whether the Settings app is allowed in kiosk mode.
"powerButtonActions": "A String", # Optional. Sets the behavior of a device in kiosk mode when a user presses and holds (long-presses) the Power button.
"statusBar": "A String", # Optional. Specifies whether system info and notifications are disabled in kiosk mode.
@@ -423,7 +423,7 @@ Method Details
},
"maximumTimeToLock": "A String", # Maximum time in milliseconds for user activity until the device locks. A value of 0 means there is no restriction.
"microphoneAccess": "A String", # Controls the use of the microphone and whether the user has access to the microphone access toggle. This applies only on fully managed devices.
- "minimumApiLevel": 42, # The minimum allowed Android API level.
+ "minimumApiLevel": 42, # The minimum allowed Android API level. A NonComplianceDetail with OS_NOT_PERMITTED is reported if the Android API level of the device is lower than this value.
"mobileNetworksConfigDisabled": True or False, # Whether configuring mobile networks is disabled.
"modifyAccountsDisabled": True or False, # Whether adding or removing accounts is disabled.
"mountPhysicalMediaDisabled": True or False, # Whether the user mounting physical external media is disabled.
@@ -944,7 +944,7 @@ Method Details
"A String",
],
"kioskCustomLauncherEnabled": True or False, # Whether the kiosk custom launcher is enabled. This replaces the home screen with a launcher that locks down the device to the apps installed via the applications setting. Apps appear on a single page in alphabetical order. Use kioskCustomization to further configure the kiosk device behavior.
- "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK.
+ "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role.
"deviceSettings": "A String", # Optional. Specifies whether the Settings app is allowed in kiosk mode.
"powerButtonActions": "A String", # Optional. Sets the behavior of a device in kiosk mode when a user presses and holds (long-presses) the Power button.
"statusBar": "A String", # Optional. Specifies whether system info and notifications are disabled in kiosk mode.
@@ -960,7 +960,7 @@ Method Details
},
"maximumTimeToLock": "A String", # Maximum time in milliseconds for user activity until the device locks. A value of 0 means there is no restriction.
"microphoneAccess": "A String", # Controls the use of the microphone and whether the user has access to the microphone access toggle. This applies only on fully managed devices.
- "minimumApiLevel": 42, # The minimum allowed Android API level.
+ "minimumApiLevel": 42, # The minimum allowed Android API level. A NonComplianceDetail with OS_NOT_PERMITTED is reported if the Android API level of the device is lower than this value.
"mobileNetworksConfigDisabled": True or False, # Whether configuring mobile networks is disabled.
"modifyAccountsDisabled": True or False, # Whether adding or removing accounts is disabled.
"mountPhysicalMediaDisabled": True or False, # Whether the user mounting physical external media is disabled.
@@ -1567,7 +1567,7 @@ Method Details
"A String",
],
"kioskCustomLauncherEnabled": True or False, # Whether the kiosk custom launcher is enabled. This replaces the home screen with a launcher that locks down the device to the apps installed via the applications setting. Apps appear on a single page in alphabetical order. Use kioskCustomization to further configure the kiosk device behavior.
- "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK.
+ "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role.
"deviceSettings": "A String", # Optional. Specifies whether the Settings app is allowed in kiosk mode.
"powerButtonActions": "A String", # Optional. Sets the behavior of a device in kiosk mode when a user presses and holds (long-presses) the Power button.
"statusBar": "A String", # Optional. Specifies whether system info and notifications are disabled in kiosk mode.
@@ -1583,7 +1583,7 @@ Method Details
},
"maximumTimeToLock": "A String", # Maximum time in milliseconds for user activity until the device locks. A value of 0 means there is no restriction.
"microphoneAccess": "A String", # Controls the use of the microphone and whether the user has access to the microphone access toggle. This applies only on fully managed devices.
- "minimumApiLevel": 42, # The minimum allowed Android API level.
+ "minimumApiLevel": 42, # The minimum allowed Android API level. A NonComplianceDetail with OS_NOT_PERMITTED is reported if the Android API level of the device is lower than this value.
"mobileNetworksConfigDisabled": True or False, # Whether configuring mobile networks is disabled.
"modifyAccountsDisabled": True or False, # Whether adding or removing accounts is disabled.
"mountPhysicalMediaDisabled": True or False, # Whether the user mounting physical external media is disabled.
@@ -2095,7 +2095,7 @@ Method Details
"A String",
],
"kioskCustomLauncherEnabled": True or False, # Whether the kiosk custom launcher is enabled. This replaces the home screen with a launcher that locks down the device to the apps installed via the applications setting. Apps appear on a single page in alphabetical order. Use kioskCustomization to further configure the kiosk device behavior.
- "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK.
+ "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role.
"deviceSettings": "A String", # Optional. Specifies whether the Settings app is allowed in kiosk mode.
"powerButtonActions": "A String", # Optional. Sets the behavior of a device in kiosk mode when a user presses and holds (long-presses) the Power button.
"statusBar": "A String", # Optional. Specifies whether system info and notifications are disabled in kiosk mode.
@@ -2111,7 +2111,7 @@ Method Details
},
"maximumTimeToLock": "A String", # Maximum time in milliseconds for user activity until the device locks. A value of 0 means there is no restriction.
"microphoneAccess": "A String", # Controls the use of the microphone and whether the user has access to the microphone access toggle. This applies only on fully managed devices.
- "minimumApiLevel": 42, # The minimum allowed Android API level.
+ "minimumApiLevel": 42, # The minimum allowed Android API level. A NonComplianceDetail with OS_NOT_PERMITTED is reported if the Android API level of the device is lower than this value.
"mobileNetworksConfigDisabled": True or False, # Whether configuring mobile networks is disabled.
"modifyAccountsDisabled": True or False, # Whether adding or removing accounts is disabled.
"mountPhysicalMediaDisabled": True or False, # Whether the user mounting physical external media is disabled.
@@ -2621,7 +2621,7 @@ Method Details
"A String",
],
"kioskCustomLauncherEnabled": True or False, # Whether the kiosk custom launcher is enabled. This replaces the home screen with a launcher that locks down the device to the apps installed via the applications setting. Apps appear on a single page in alphabetical order. Use kioskCustomization to further configure the kiosk device behavior.
- "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK.
+ "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role.
"deviceSettings": "A String", # Optional. Specifies whether the Settings app is allowed in kiosk mode.
"powerButtonActions": "A String", # Optional. Sets the behavior of a device in kiosk mode when a user presses and holds (long-presses) the Power button.
"statusBar": "A String", # Optional. Specifies whether system info and notifications are disabled in kiosk mode.
@@ -2637,7 +2637,7 @@ Method Details
},
"maximumTimeToLock": "A String", # Maximum time in milliseconds for user activity until the device locks. A value of 0 means there is no restriction.
"microphoneAccess": "A String", # Controls the use of the microphone and whether the user has access to the microphone access toggle. This applies only on fully managed devices.
- "minimumApiLevel": 42, # The minimum allowed Android API level.
+ "minimumApiLevel": 42, # The minimum allowed Android API level. A NonComplianceDetail with OS_NOT_PERMITTED is reported if the Android API level of the device is lower than this value.
"mobileNetworksConfigDisabled": True or False, # Whether configuring mobile networks is disabled.
"modifyAccountsDisabled": True or False, # Whether adding or removing accounts is disabled.
"mountPhysicalMediaDisabled": True or False, # Whether the user mounting physical external media is disabled.
@@ -3163,7 +3163,7 @@ Method Details
"A String",
],
"kioskCustomLauncherEnabled": True or False, # Whether the kiosk custom launcher is enabled. This replaces the home screen with a launcher that locks down the device to the apps installed via the applications setting. Apps appear on a single page in alphabetical order. Use kioskCustomization to further configure the kiosk device behavior.
- "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with installType KIOSK.
+ "kioskCustomization": { # Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role. # Optional. Settings controlling the behavior of a device in kiosk mode. To enable kiosk mode, set kioskCustomLauncherEnabled to true or specify an app in the policy with the KIOSK role.
"deviceSettings": "A String", # Optional. Specifies whether the Settings app is allowed in kiosk mode.
"powerButtonActions": "A String", # Optional. Sets the behavior of a device in kiosk mode when a user presses and holds (long-presses) the Power button.
"statusBar": "A String", # Optional. Specifies whether system info and notifications are disabled in kiosk mode.
@@ -3179,7 +3179,7 @@ Method Details
},
"maximumTimeToLock": "A String", # Maximum time in milliseconds for user activity until the device locks. A value of 0 means there is no restriction.
"microphoneAccess": "A String", # Controls the use of the microphone and whether the user has access to the microphone access toggle. This applies only on fully managed devices.
- "minimumApiLevel": 42, # The minimum allowed Android API level.
+ "minimumApiLevel": 42, # The minimum allowed Android API level. A NonComplianceDetail with OS_NOT_PERMITTED is reported if the Android API level of the device is lower than this value.
"mobileNetworksConfigDisabled": True or False, # Whether configuring mobile networks is disabled.
"modifyAccountsDisabled": True or False, # Whether adding or removing accounts is disabled.
"mountPhysicalMediaDisabled": True or False, # Whether the user mounting physical external media is disabled.
diff --git a/docs/dyn/apigee_v1.organizations.html b/docs/dyn/apigee_v1.organizations.html
index d40800675e..176cd5d4dd 100644
--- a/docs/dyn/apigee_v1.organizations.html
+++ b/docs/dyn/apigee_v1.organizations.html
@@ -394,6 +394,7 @@ Method Details
Allowed values
DELETION_RETENTION_UNSPECIFIED - Default data retention setting of seven days will be applied.
MINIMUM - Organization data will be retained for the minimum period of 24 hours.
+ NO_RETENTION - Does not retain organization data and permanently deletes the organization immediately. **Warning:** You will NOT be able to restore the Apigee organization when using this option.
x__xgafv: string, V1 error format.
Allowed values
1 - v1 error format
diff --git a/docs/dyn/artifactregistry_v1.projects.locations.repositories.html b/docs/dyn/artifactregistry_v1.projects.locations.repositories.html
index 5a6d52186b..d3a51fb9e3 100644
--- a/docs/dyn/artifactregistry_v1.projects.locations.repositories.html
+++ b/docs/dyn/artifactregistry_v1.projects.locations.repositories.html
@@ -291,7 +291,7 @@ Method Details
"remoteRepositoryConfig": { # Remote repository configuration. # Configuration specific for a Remote Repository.
"aptRepository": { # Configuration for an Apt remote repository. # Specific settings for an Apt remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
},
"publicRepository": { # Publicly available Apt repositories constructed from a common repository base and a custom repository path. # One of the publicly available Apt repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Apt.
@@ -305,13 +305,13 @@ Method Details
"disableUpstreamValidation": True or False, # Input only. A create/update remote repo option to avoid making a HEAD/GET request to validate a remote repo and any supplied upstream credentials.
"dockerRepository": { # Configuration for a Docker remote repository. # Specific settings for a Docker remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
+ "uri": "A String", # An https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
},
"publicRepository": "A String", # One of the publicly available Docker repositories supported by Artifact Registry.
},
"mavenRepository": { # Configuration for a Maven remote repository. # Specific settings for a Maven remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
},
"publicRepository": "A String", # One of the publicly available Maven repositories supported by Artifact Registry.
},
@@ -319,13 +319,13 @@ Method Details
},
"npmRepository": { # Configuration for a Npm remote repository. # Specific settings for an Npm remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
},
"publicRepository": "A String", # One of the publicly available Npm repositories supported by Artifact Registry.
},
"pythonRepository": { # Configuration for a Python remote repository. # Specific settings for a Python remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
},
"publicRepository": "A String", # One of the publicly available Python repositories supported by Artifact Registry.
},
@@ -337,7 +337,7 @@ Method Details
},
"yumRepository": { # Configuration for a Yum remote repository. # Specific settings for a Yum remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
},
"publicRepository": { # Publicly available Yum repositories constructed from a common repository base and a custom repository path. # One of the publicly available Yum repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Yum.
@@ -542,7 +542,7 @@ Method Details
"remoteRepositoryConfig": { # Remote repository configuration. # Configuration specific for a Remote Repository.
"aptRepository": { # Configuration for an Apt remote repository. # Specific settings for an Apt remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
},
"publicRepository": { # Publicly available Apt repositories constructed from a common repository base and a custom repository path. # One of the publicly available Apt repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Apt.
@@ -556,13 +556,13 @@ Method Details
"disableUpstreamValidation": True or False, # Input only. A create/update remote repo option to avoid making a HEAD/GET request to validate a remote repo and any supplied upstream credentials.
"dockerRepository": { # Configuration for a Docker remote repository. # Specific settings for a Docker remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
+ "uri": "A String", # An https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
},
"publicRepository": "A String", # One of the publicly available Docker repositories supported by Artifact Registry.
},
"mavenRepository": { # Configuration for a Maven remote repository. # Specific settings for a Maven remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
},
"publicRepository": "A String", # One of the publicly available Maven repositories supported by Artifact Registry.
},
@@ -570,13 +570,13 @@ Method Details
},
"npmRepository": { # Configuration for a Npm remote repository. # Specific settings for an Npm remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
},
"publicRepository": "A String", # One of the publicly available Npm repositories supported by Artifact Registry.
},
"pythonRepository": { # Configuration for a Python remote repository. # Specific settings for a Python remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
},
"publicRepository": "A String", # One of the publicly available Python repositories supported by Artifact Registry.
},
@@ -588,7 +588,7 @@ Method Details
},
"yumRepository": { # Configuration for a Yum remote repository. # Specific settings for a Yum remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
},
"publicRepository": { # Publicly available Yum repositories constructed from a common repository base and a custom repository path. # One of the publicly available Yum repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Yum.
@@ -727,7 +727,7 @@ Method Details
"remoteRepositoryConfig": { # Remote repository configuration. # Configuration specific for a Remote Repository.
"aptRepository": { # Configuration for an Apt remote repository. # Specific settings for an Apt remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
},
"publicRepository": { # Publicly available Apt repositories constructed from a common repository base and a custom repository path. # One of the publicly available Apt repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Apt.
@@ -741,13 +741,13 @@ Method Details
"disableUpstreamValidation": True or False, # Input only. A create/update remote repo option to avoid making a HEAD/GET request to validate a remote repo and any supplied upstream credentials.
"dockerRepository": { # Configuration for a Docker remote repository. # Specific settings for a Docker remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
+ "uri": "A String", # An https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
},
"publicRepository": "A String", # One of the publicly available Docker repositories supported by Artifact Registry.
},
"mavenRepository": { # Configuration for a Maven remote repository. # Specific settings for a Maven remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
},
"publicRepository": "A String", # One of the publicly available Maven repositories supported by Artifact Registry.
},
@@ -755,13 +755,13 @@ Method Details
},
"npmRepository": { # Configuration for a Npm remote repository. # Specific settings for an Npm remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
},
"publicRepository": "A String", # One of the publicly available Npm repositories supported by Artifact Registry.
},
"pythonRepository": { # Configuration for a Python remote repository. # Specific settings for a Python remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
},
"publicRepository": "A String", # One of the publicly available Python repositories supported by Artifact Registry.
},
@@ -773,7 +773,7 @@ Method Details
},
"yumRepository": { # Configuration for a Yum remote repository. # Specific settings for a Yum remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
},
"publicRepository": { # Publicly available Yum repositories constructed from a common repository base and a custom repository path. # One of the publicly available Yum repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Yum.
@@ -881,7 +881,7 @@ Method Details
"remoteRepositoryConfig": { # Remote repository configuration. # Configuration specific for a Remote Repository.
"aptRepository": { # Configuration for an Apt remote repository. # Specific settings for an Apt remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
},
"publicRepository": { # Publicly available Apt repositories constructed from a common repository base and a custom repository path. # One of the publicly available Apt repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Apt.
@@ -895,13 +895,13 @@ Method Details
"disableUpstreamValidation": True or False, # Input only. A create/update remote repo option to avoid making a HEAD/GET request to validate a remote repo and any supplied upstream credentials.
"dockerRepository": { # Configuration for a Docker remote repository. # Specific settings for a Docker remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
+ "uri": "A String", # An https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
},
"publicRepository": "A String", # One of the publicly available Docker repositories supported by Artifact Registry.
},
"mavenRepository": { # Configuration for a Maven remote repository. # Specific settings for a Maven remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
},
"publicRepository": "A String", # One of the publicly available Maven repositories supported by Artifact Registry.
},
@@ -909,13 +909,13 @@ Method Details
},
"npmRepository": { # Configuration for a Npm remote repository. # Specific settings for an Npm remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
},
"publicRepository": "A String", # One of the publicly available Npm repositories supported by Artifact Registry.
},
"pythonRepository": { # Configuration for a Python remote repository. # Specific settings for a Python remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
},
"publicRepository": "A String", # One of the publicly available Python repositories supported by Artifact Registry.
},
@@ -927,7 +927,7 @@ Method Details
},
"yumRepository": { # Configuration for a Yum remote repository. # Specific settings for a Yum remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
},
"publicRepository": { # Publicly available Yum repositories constructed from a common repository base and a custom repository path. # One of the publicly available Yum repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Yum.
@@ -1018,7 +1018,7 @@ Method Details
"remoteRepositoryConfig": { # Remote repository configuration. # Configuration specific for a Remote Repository.
"aptRepository": { # Configuration for an Apt remote repository. # Specific settings for an Apt remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.apt.registry/".
},
"publicRepository": { # Publicly available Apt repositories constructed from a common repository base and a custom repository path. # One of the publicly available Apt repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Apt.
@@ -1032,13 +1032,13 @@ Method Details
"disableUpstreamValidation": True or False, # Input only. A create/update remote repo option to avoid making a HEAD/GET request to validate a remote repo and any supplied upstream credentials.
"dockerRepository": { # Configuration for a Docker remote repository. # Specific settings for a Docker remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
+ "uri": "A String", # An https uri reference to the custom remote repository, for ex: "https://registry-1.docker.io".
},
"publicRepository": "A String", # One of the publicly available Docker repositories supported by Artifact Registry.
},
"mavenRepository": { # Configuration for a Maven remote repository. # Specific settings for a Maven remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.maven.registry/".
},
"publicRepository": "A String", # One of the publicly available Maven repositories supported by Artifact Registry.
},
@@ -1046,13 +1046,13 @@ Method Details
},
"npmRepository": { # Configuration for a Npm remote repository. # Specific settings for an Npm remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.npm.registry/".
},
"publicRepository": "A String", # One of the publicly available Npm repositories supported by Artifact Registry.
},
"pythonRepository": { # Configuration for a Python remote repository. # Specific settings for a Python remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.python.registry/".
},
"publicRepository": "A String", # One of the publicly available Python repositories supported by Artifact Registry.
},
@@ -1064,7 +1064,7 @@ Method Details
},
"yumRepository": { # Configuration for a Yum remote repository. # Specific settings for a Yum remote repository.
"customRepository": { # Customer-specified publicly available remote repository. # Customer-specified remote repository.
- "uri": "A String", # An http/https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
+ "uri": "A String", # An https uri reference to the upstream remote repository, for ex: "https://my.yum.registry/".
},
"publicRepository": { # Publicly available Yum repositories constructed from a common repository base and a custom repository path. # One of the publicly available Yum repositories supported by Artifact Registry.
"repositoryBase": "A String", # A common public repository base for Yum.
diff --git a/docs/dyn/assuredworkloads_v1.html b/docs/dyn/assuredworkloads_v1.html
index 6fecf0415a..fea59404d9 100644
--- a/docs/dyn/assuredworkloads_v1.html
+++ b/docs/dyn/assuredworkloads_v1.html
@@ -74,11 +74,6 @@
Assured Workloads API
Instance Methods
-
-Returns the assuredworkloads Resource.
-
diff --git a/docs/dyn/assuredworkloads_v1beta1.html b/docs/dyn/assuredworkloads_v1beta1.html
index 075e3b296d..ed031cf364 100644
--- a/docs/dyn/assuredworkloads_v1beta1.html
+++ b/docs/dyn/assuredworkloads_v1beta1.html
@@ -74,11 +74,6 @@
Assured Workloads API
Instance Methods
-
-Returns the assuredworkloads Resource.
-
diff --git a/docs/dyn/biglake_v1.projects.catalogs.namespaces.html b/docs/dyn/biglake_v1.projects.catalogs.namespaces.html
index c953d57710..b88bb9de04 100644
--- a/docs/dyn/biglake_v1.projects.catalogs.namespaces.html
+++ b/docs/dyn/biglake_v1.projects.catalogs.namespaces.html
@@ -79,6 +79,11 @@ Instance Methods
Returns the tables Resource.
+
+ views()
+
+Returns the views Resource.
+
Close httplib2 connections.
diff --git a/docs/dyn/biglake_v1.projects.catalogs.namespaces.views.html b/docs/dyn/biglake_v1.projects.catalogs.namespaces.views.html
new file mode 100644
index 0000000000..74adaa9dac
--- /dev/null
+++ b/docs/dyn/biglake_v1.projects.catalogs.namespaces.views.html
@@ -0,0 +1,258 @@
+
+
+
+Lakehouse API . projects . catalogs . namespaces . views
+Instance Methods
+
+ close()
+Close httplib2 connections.
+
+ getIamPolicy(resource, options_requestedPolicyVersion=None, x__xgafv=None)
+Gets the IAM policy for the specified View.
+
+ setIamPolicy(resource, body=None, x__xgafv=None)
+Sets the IAM policy for the specified view.
+
+ testIamPermissions(resource, body=None, x__xgafv=None)
+Tests the IAM permissions for the specified view.
+Method Details
+
+ close()
+ Close httplib2 connections.
+
+
+
+ getIamPolicy(resource, options_requestedPolicyVersion=None, x__xgafv=None)
+ Gets the IAM policy for the specified View.
+
+Args:
+ resource: string, REQUIRED: The resource for which the policy is being requested. See [Resource names](https://cloud.google.com/apis/design/resource_names) for the appropriate value for this field. (required)
+ options_requestedPolicyVersion: integer, Optional. The maximum policy version that will be used to format the policy. Valid values are 0, 1, and 3. Requests specifying an invalid value will be rejected. Requests for policies with any conditional role bindings must specify version 3. Policies with no conditional role bindings may specify any valid value or leave the field unset. The policy in the response might use the policy version that you specified, or it might use a lower policy version. For example, if you specify version 3, but the policy has no conditional role bindings, the response uses version 1. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # An Identity and Access Management (IAM) policy, which specifies access controls for Google Cloud resources. A `Policy` is a collection of `bindings`. A `binding` binds one or more `members`, or principals, to a single `role`. Principals can be user accounts, service accounts, Google groups, and domains (such as G Suite). A `role` is a named list of permissions; each `role` can be an IAM predefined role or a user-created custom role. For some types of Google Cloud resources, a `binding` can also specify a `condition`, which is a logical expression that allows access to a resource only if the expression evaluates to `true`. A condition can add constraints based on attributes of the request, the resource, or both. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies). **JSON example:** ``` { "bindings": [ { "role": "roles/resourcemanager.organizationAdmin", "members": [ "user:mike@example.com", "group:admins@example.com", "domain:google.com", "serviceAccount:my-project-id@appspot.gserviceaccount.com" ] }, { "role": "roles/resourcemanager.organizationViewer", "members": [ "user:eve@example.com" ], "condition": { "title": "expirable access", "description": "Does not grant access after Sep 2020", "expression": "request.time < timestamp('2020-10-01T00:00:00.000Z')", } } ], "etag": "BwWWja0YfJA=", "version": 3 } ``` **YAML example:** ``` bindings: - members: - user:mike@example.com - group:admins@example.com - domain:google.com - serviceAccount:my-project-id@appspot.gserviceaccount.com role: roles/resourcemanager.organizationAdmin - members: - user:eve@example.com role: roles/resourcemanager.organizationViewer condition: title: expirable access description: Does not grant access after Sep 2020 expression: request.time < timestamp('2020-10-01T00:00:00.000Z') etag: BwWWja0YfJA= version: 3 ``` For a description of IAM and its features, see the [IAM documentation](https://cloud.google.com/iam/docs/).
+ "auditConfigs": [ # Specifies cloud audit logging configuration for this policy.
+ { # Specifies the audit configuration for a service. The configuration determines which permission types are logged, and what identities, if any, are exempted from logging. An AuditConfig must have one or more AuditLogConfigs. If there are AuditConfigs for both `allServices` and a specific service, the union of the two AuditConfigs is used for that service: the log_types specified in each AuditConfig are enabled, and the exempted_members in each AuditLogConfig are exempted. Example Policy with multiple AuditConfigs: { "audit_configs": [ { "service": "allServices", "audit_log_configs": [ { "log_type": "DATA_READ", "exempted_members": [ "user:jose@example.com" ] }, { "log_type": "DATA_WRITE" }, { "log_type": "ADMIN_READ" } ] }, { "service": "sampleservice.googleapis.com", "audit_log_configs": [ { "log_type": "DATA_READ" }, { "log_type": "DATA_WRITE", "exempted_members": [ "user:aliya@example.com" ] } ] } ] } For sampleservice, this policy enables DATA_READ, DATA_WRITE and ADMIN_READ logging. It also exempts `jose@example.com` from DATA_READ logging, and `aliya@example.com` from DATA_WRITE logging.
+ "auditLogConfigs": [ # The configuration for logging of each type of permission.
+ { # Provides the configuration for logging a type of permissions. Example: { "audit_log_configs": [ { "log_type": "DATA_READ", "exempted_members": [ "user:jose@example.com" ] }, { "log_type": "DATA_WRITE" } ] } This enables 'DATA_READ' and 'DATA_WRITE' logging, while exempting jose@example.com from DATA_READ logging.
+ "exemptedMembers": [ # Specifies the identities that do not cause logging for this type of permission. Follows the same format of Binding.members.
+ "A String",
+ ],
+ "logType": "A String", # The log type that this config enables.
+ },
+ ],
+ "service": "A String", # Specifies a service that will be enabled for audit logging. For example, `storage.googleapis.com`, `cloudsql.googleapis.com`. `allServices` is a special value that covers all services.
+ },
+ ],
+ "bindings": [ # Associates a list of `members`, or principals, with a `role`. Optionally, may specify a `condition` that determines how and when the `bindings` are applied. Each of the `bindings` must contain at least one principal. The `bindings` in a `Policy` can refer to up to 1,500 principals; up to 250 of these principals can be Google groups. Each occurrence of a principal counts towards these limits. For example, if the `bindings` grant 50 different roles to `user:alice@example.com`, and not to any other principal, then you can add another 1,450 principals to the `bindings` in the `Policy`.
+ { # Associates `members`, or principals, with a `role`.
+ "condition": { # Represents a textual expression in the Common Expression Language (CEL) syntax. CEL is a C-like expression language. The syntax and semantics of CEL are documented at https://github.com/google/cel-spec. Example (Comparison): title: "Summary size limit" description: "Determines if a summary is less than 100 chars" expression: "document.summary.size() < 100" Example (Equality): title: "Requestor is owner" description: "Determines if requestor is the document owner" expression: "document.owner == request.auth.claims.email" Example (Logic): title: "Public documents" description: "Determine whether the document should be publicly visible" expression: "document.type != 'private' && document.type != 'internal'" Example (Data Manipulation): title: "Notification string" description: "Create a notification string with a timestamp." expression: "'New message received at ' + string(document.create_time)" The exact variables and functions that may be referenced within an expression are determined by the service that evaluates it. See the service documentation for additional information. # The condition that is associated with this binding. If the condition evaluates to `true`, then this binding applies to the current request. If the condition evaluates to `false`, then this binding does not apply to the current request. However, a different role binding might grant the same role to one or more of the principals in this binding. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+ "description": "A String", # Optional. Description of the expression. This is a longer text which describes the expression, e.g. when hovered over it in a UI.
+ "expression": "A String", # Textual representation of an expression in Common Expression Language syntax.
+ "location": "A String", # Optional. String indicating the location of the expression for error reporting, e.g. a file name and a position in the file.
+ "title": "A String", # Optional. Title for the expression, i.e. a short string describing its purpose. This can be used e.g. in UIs which allow to enter the expression.
+ },
+ "members": [ # Specifies the principals requesting access for a Google Cloud resource. `members` can have the following values: * `allUsers`: A special identifier that represents anyone who is on the internet; with or without a Google account. * `allAuthenticatedUsers`: A special identifier that represents anyone who is authenticated with a Google account or a service account. Does not include identities that come from external identity providers (IdPs) through identity federation. * `user:{emailid}`: An email address that represents a specific Google account. For example, `alice@example.com` . * `serviceAccount:{emailid}`: An email address that represents a Google service account. For example, `my-other-app@appspot.gserviceaccount.com`. * `serviceAccount:{projectid}.svc.id.goog[{namespace}/{kubernetes-sa}]`: An identifier for a [Kubernetes service account](https://cloud.google.com/kubernetes-engine/docs/how-to/kubernetes-service-accounts). For example, `my-project.svc.id.goog[my-namespace/my-kubernetes-sa]`. * `group:{emailid}`: An email address that represents a Google group. For example, `admins@example.com`. * `domain:{domain}`: The G Suite domain (primary) that represents all the users of that domain. For example, `google.com` or `example.com`. * `principal://iam.googleapis.com/locations/global/workforcePools/{pool_id}/subject/{subject_attribute_value}`: A single identity in a workforce identity pool. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/group/{group_id}`: All workforce identities in a group. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/attribute.{attribute_name}/{attribute_value}`: All workforce identities with a specific attribute value. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/*`: All identities in a workforce identity pool. * `principal://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/subject/{subject_attribute_value}`: A single identity in a workload identity pool. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/group/{group_id}`: A workload identity pool group. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/attribute.{attribute_name}/{attribute_value}`: All identities in a workload identity pool with a certain attribute. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/*`: All identities in a workload identity pool. * `deleted:user:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a user that has been recently deleted. For example, `alice@example.com?uid=123456789012345678901`. If the user is recovered, this value reverts to `user:{emailid}` and the recovered user retains the role in the binding. * `deleted:serviceAccount:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a service account that has been recently deleted. For example, `my-other-app@appspot.gserviceaccount.com?uid=123456789012345678901`. If the service account is undeleted, this value reverts to `serviceAccount:{emailid}` and the undeleted service account retains the role in the binding. * `deleted:group:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a Google group that has been recently deleted. For example, `admins@example.com?uid=123456789012345678901`. If the group is recovered, this value reverts to `group:{emailid}` and the recovered group retains the role in the binding. * `deleted:principal://iam.googleapis.com/locations/global/workforcePools/{pool_id}/subject/{subject_attribute_value}`: Deleted single identity in a workforce identity pool. For example, `deleted:principal://iam.googleapis.com/locations/global/workforcePools/my-pool-id/subject/my-subject-attribute-value`.
+ "A String",
+ ],
+ "role": "A String", # Role that is assigned to the list of `members`, or principals. For example, `roles/viewer`, `roles/editor`, or `roles/owner`. For an overview of the IAM roles and permissions, see the [IAM documentation](https://cloud.google.com/iam/docs/roles-overview). For a list of the available pre-defined roles, see [here](https://cloud.google.com/iam/docs/understanding-roles).
+ },
+ ],
+ "etag": "A String", # `etag` is used for optimistic concurrency control as a way to help prevent simultaneous updates of a policy from overwriting each other. It is strongly suggested that systems make use of the `etag` in the read-modify-write cycle to perform policy updates in order to avoid race conditions: An `etag` is returned in the response to `getIamPolicy`, and systems are expected to put that etag in the request to `setIamPolicy` to ensure that their change will be applied to the same version of the policy. **Important:** If you use IAM Conditions, you must include the `etag` field whenever you call `setIamPolicy`. If you omit this field, then IAM allows you to overwrite a version `3` policy with a version `1` policy, and all of the conditions in the version `3` policy are lost.
+ "version": 42, # Specifies the format of the policy. Valid values are `0`, `1`, and `3`. Requests that specify an invalid value are rejected. Any operation that affects conditional role bindings must specify version `3`. This requirement applies to the following operations: * Getting a policy that includes a conditional role binding * Adding a conditional role binding to a policy * Changing a conditional role binding in a policy * Removing any role binding, with or without a condition, from a policy that includes conditions **Important:** If you use IAM Conditions, you must include the `etag` field whenever you call `setIamPolicy`. If you omit this field, then IAM allows you to overwrite a version `3` policy with a version `1` policy, and all of the conditions in the version `3` policy are lost. If a policy does not include any conditions, operations on that policy may specify any valid version or leave the field unset. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+}
+
+
+
+ setIamPolicy(resource, body=None, x__xgafv=None)
+ Sets the IAM policy for the specified view.
+
+Args:
+ resource: string, REQUIRED: The resource for which the policy is being specified. See [Resource names](https://cloud.google.com/apis/design/resource_names) for the appropriate value for this field. (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request message for `SetIamPolicy` method.
+ "policy": { # An Identity and Access Management (IAM) policy, which specifies access controls for Google Cloud resources. A `Policy` is a collection of `bindings`. A `binding` binds one or more `members`, or principals, to a single `role`. Principals can be user accounts, service accounts, Google groups, and domains (such as G Suite). A `role` is a named list of permissions; each `role` can be an IAM predefined role or a user-created custom role. For some types of Google Cloud resources, a `binding` can also specify a `condition`, which is a logical expression that allows access to a resource only if the expression evaluates to `true`. A condition can add constraints based on attributes of the request, the resource, or both. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies). **JSON example:** ``` { "bindings": [ { "role": "roles/resourcemanager.organizationAdmin", "members": [ "user:mike@example.com", "group:admins@example.com", "domain:google.com", "serviceAccount:my-project-id@appspot.gserviceaccount.com" ] }, { "role": "roles/resourcemanager.organizationViewer", "members": [ "user:eve@example.com" ], "condition": { "title": "expirable access", "description": "Does not grant access after Sep 2020", "expression": "request.time < timestamp('2020-10-01T00:00:00.000Z')", } } ], "etag": "BwWWja0YfJA=", "version": 3 } ``` **YAML example:** ``` bindings: - members: - user:mike@example.com - group:admins@example.com - domain:google.com - serviceAccount:my-project-id@appspot.gserviceaccount.com role: roles/resourcemanager.organizationAdmin - members: - user:eve@example.com role: roles/resourcemanager.organizationViewer condition: title: expirable access description: Does not grant access after Sep 2020 expression: request.time < timestamp('2020-10-01T00:00:00.000Z') etag: BwWWja0YfJA= version: 3 ``` For a description of IAM and its features, see the [IAM documentation](https://cloud.google.com/iam/docs/). # REQUIRED: The complete policy to be applied to the `resource`. The size of the policy is limited to a few 10s of KB. An empty policy is a valid policy but certain Google Cloud services (such as Projects) might reject them.
+ "auditConfigs": [ # Specifies cloud audit logging configuration for this policy.
+ { # Specifies the audit configuration for a service. The configuration determines which permission types are logged, and what identities, if any, are exempted from logging. An AuditConfig must have one or more AuditLogConfigs. If there are AuditConfigs for both `allServices` and a specific service, the union of the two AuditConfigs is used for that service: the log_types specified in each AuditConfig are enabled, and the exempted_members in each AuditLogConfig are exempted. Example Policy with multiple AuditConfigs: { "audit_configs": [ { "service": "allServices", "audit_log_configs": [ { "log_type": "DATA_READ", "exempted_members": [ "user:jose@example.com" ] }, { "log_type": "DATA_WRITE" }, { "log_type": "ADMIN_READ" } ] }, { "service": "sampleservice.googleapis.com", "audit_log_configs": [ { "log_type": "DATA_READ" }, { "log_type": "DATA_WRITE", "exempted_members": [ "user:aliya@example.com" ] } ] } ] } For sampleservice, this policy enables DATA_READ, DATA_WRITE and ADMIN_READ logging. It also exempts `jose@example.com` from DATA_READ logging, and `aliya@example.com` from DATA_WRITE logging.
+ "auditLogConfigs": [ # The configuration for logging of each type of permission.
+ { # Provides the configuration for logging a type of permissions. Example: { "audit_log_configs": [ { "log_type": "DATA_READ", "exempted_members": [ "user:jose@example.com" ] }, { "log_type": "DATA_WRITE" } ] } This enables 'DATA_READ' and 'DATA_WRITE' logging, while exempting jose@example.com from DATA_READ logging.
+ "exemptedMembers": [ # Specifies the identities that do not cause logging for this type of permission. Follows the same format of Binding.members.
+ "A String",
+ ],
+ "logType": "A String", # The log type that this config enables.
+ },
+ ],
+ "service": "A String", # Specifies a service that will be enabled for audit logging. For example, `storage.googleapis.com`, `cloudsql.googleapis.com`. `allServices` is a special value that covers all services.
+ },
+ ],
+ "bindings": [ # Associates a list of `members`, or principals, with a `role`. Optionally, may specify a `condition` that determines how and when the `bindings` are applied. Each of the `bindings` must contain at least one principal. The `bindings` in a `Policy` can refer to up to 1,500 principals; up to 250 of these principals can be Google groups. Each occurrence of a principal counts towards these limits. For example, if the `bindings` grant 50 different roles to `user:alice@example.com`, and not to any other principal, then you can add another 1,450 principals to the `bindings` in the `Policy`.
+ { # Associates `members`, or principals, with a `role`.
+ "condition": { # Represents a textual expression in the Common Expression Language (CEL) syntax. CEL is a C-like expression language. The syntax and semantics of CEL are documented at https://github.com/google/cel-spec. Example (Comparison): title: "Summary size limit" description: "Determines if a summary is less than 100 chars" expression: "document.summary.size() < 100" Example (Equality): title: "Requestor is owner" description: "Determines if requestor is the document owner" expression: "document.owner == request.auth.claims.email" Example (Logic): title: "Public documents" description: "Determine whether the document should be publicly visible" expression: "document.type != 'private' && document.type != 'internal'" Example (Data Manipulation): title: "Notification string" description: "Create a notification string with a timestamp." expression: "'New message received at ' + string(document.create_time)" The exact variables and functions that may be referenced within an expression are determined by the service that evaluates it. See the service documentation for additional information. # The condition that is associated with this binding. If the condition evaluates to `true`, then this binding applies to the current request. If the condition evaluates to `false`, then this binding does not apply to the current request. However, a different role binding might grant the same role to one or more of the principals in this binding. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+ "description": "A String", # Optional. Description of the expression. This is a longer text which describes the expression, e.g. when hovered over it in a UI.
+ "expression": "A String", # Textual representation of an expression in Common Expression Language syntax.
+ "location": "A String", # Optional. String indicating the location of the expression for error reporting, e.g. a file name and a position in the file.
+ "title": "A String", # Optional. Title for the expression, i.e. a short string describing its purpose. This can be used e.g. in UIs which allow to enter the expression.
+ },
+ "members": [ # Specifies the principals requesting access for a Google Cloud resource. `members` can have the following values: * `allUsers`: A special identifier that represents anyone who is on the internet; with or without a Google account. * `allAuthenticatedUsers`: A special identifier that represents anyone who is authenticated with a Google account or a service account. Does not include identities that come from external identity providers (IdPs) through identity federation. * `user:{emailid}`: An email address that represents a specific Google account. For example, `alice@example.com` . * `serviceAccount:{emailid}`: An email address that represents a Google service account. For example, `my-other-app@appspot.gserviceaccount.com`. * `serviceAccount:{projectid}.svc.id.goog[{namespace}/{kubernetes-sa}]`: An identifier for a [Kubernetes service account](https://cloud.google.com/kubernetes-engine/docs/how-to/kubernetes-service-accounts). For example, `my-project.svc.id.goog[my-namespace/my-kubernetes-sa]`. * `group:{emailid}`: An email address that represents a Google group. For example, `admins@example.com`. * `domain:{domain}`: The G Suite domain (primary) that represents all the users of that domain. For example, `google.com` or `example.com`. * `principal://iam.googleapis.com/locations/global/workforcePools/{pool_id}/subject/{subject_attribute_value}`: A single identity in a workforce identity pool. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/group/{group_id}`: All workforce identities in a group. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/attribute.{attribute_name}/{attribute_value}`: All workforce identities with a specific attribute value. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/*`: All identities in a workforce identity pool. * `principal://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/subject/{subject_attribute_value}`: A single identity in a workload identity pool. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/group/{group_id}`: A workload identity pool group. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/attribute.{attribute_name}/{attribute_value}`: All identities in a workload identity pool with a certain attribute. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/*`: All identities in a workload identity pool. * `deleted:user:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a user that has been recently deleted. For example, `alice@example.com?uid=123456789012345678901`. If the user is recovered, this value reverts to `user:{emailid}` and the recovered user retains the role in the binding. * `deleted:serviceAccount:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a service account that has been recently deleted. For example, `my-other-app@appspot.gserviceaccount.com?uid=123456789012345678901`. If the service account is undeleted, this value reverts to `serviceAccount:{emailid}` and the undeleted service account retains the role in the binding. * `deleted:group:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a Google group that has been recently deleted. For example, `admins@example.com?uid=123456789012345678901`. If the group is recovered, this value reverts to `group:{emailid}` and the recovered group retains the role in the binding. * `deleted:principal://iam.googleapis.com/locations/global/workforcePools/{pool_id}/subject/{subject_attribute_value}`: Deleted single identity in a workforce identity pool. For example, `deleted:principal://iam.googleapis.com/locations/global/workforcePools/my-pool-id/subject/my-subject-attribute-value`.
+ "A String",
+ ],
+ "role": "A String", # Role that is assigned to the list of `members`, or principals. For example, `roles/viewer`, `roles/editor`, or `roles/owner`. For an overview of the IAM roles and permissions, see the [IAM documentation](https://cloud.google.com/iam/docs/roles-overview). For a list of the available pre-defined roles, see [here](https://cloud.google.com/iam/docs/understanding-roles).
+ },
+ ],
+ "etag": "A String", # `etag` is used for optimistic concurrency control as a way to help prevent simultaneous updates of a policy from overwriting each other. It is strongly suggested that systems make use of the `etag` in the read-modify-write cycle to perform policy updates in order to avoid race conditions: An `etag` is returned in the response to `getIamPolicy`, and systems are expected to put that etag in the request to `setIamPolicy` to ensure that their change will be applied to the same version of the policy. **Important:** If you use IAM Conditions, you must include the `etag` field whenever you call `setIamPolicy`. If you omit this field, then IAM allows you to overwrite a version `3` policy with a version `1` policy, and all of the conditions in the version `3` policy are lost.
+ "version": 42, # Specifies the format of the policy. Valid values are `0`, `1`, and `3`. Requests that specify an invalid value are rejected. Any operation that affects conditional role bindings must specify version `3`. This requirement applies to the following operations: * Getting a policy that includes a conditional role binding * Adding a conditional role binding to a policy * Changing a conditional role binding in a policy * Removing any role binding, with or without a condition, from a policy that includes conditions **Important:** If you use IAM Conditions, you must include the `etag` field whenever you call `setIamPolicy`. If you omit this field, then IAM allows you to overwrite a version `3` policy with a version `1` policy, and all of the conditions in the version `3` policy are lost. If a policy does not include any conditions, operations on that policy may specify any valid version or leave the field unset. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+ },
+ "updateMask": "A String", # OPTIONAL: A FieldMask specifying which fields of the policy to modify. Only the fields in the mask will be modified. If no mask is provided, the following default mask is used: `paths: "bindings, etag"`
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # An Identity and Access Management (IAM) policy, which specifies access controls for Google Cloud resources. A `Policy` is a collection of `bindings`. A `binding` binds one or more `members`, or principals, to a single `role`. Principals can be user accounts, service accounts, Google groups, and domains (such as G Suite). A `role` is a named list of permissions; each `role` can be an IAM predefined role or a user-created custom role. For some types of Google Cloud resources, a `binding` can also specify a `condition`, which is a logical expression that allows access to a resource only if the expression evaluates to `true`. A condition can add constraints based on attributes of the request, the resource, or both. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies). **JSON example:** ``` { "bindings": [ { "role": "roles/resourcemanager.organizationAdmin", "members": [ "user:mike@example.com", "group:admins@example.com", "domain:google.com", "serviceAccount:my-project-id@appspot.gserviceaccount.com" ] }, { "role": "roles/resourcemanager.organizationViewer", "members": [ "user:eve@example.com" ], "condition": { "title": "expirable access", "description": "Does not grant access after Sep 2020", "expression": "request.time < timestamp('2020-10-01T00:00:00.000Z')", } } ], "etag": "BwWWja0YfJA=", "version": 3 } ``` **YAML example:** ``` bindings: - members: - user:mike@example.com - group:admins@example.com - domain:google.com - serviceAccount:my-project-id@appspot.gserviceaccount.com role: roles/resourcemanager.organizationAdmin - members: - user:eve@example.com role: roles/resourcemanager.organizationViewer condition: title: expirable access description: Does not grant access after Sep 2020 expression: request.time < timestamp('2020-10-01T00:00:00.000Z') etag: BwWWja0YfJA= version: 3 ``` For a description of IAM and its features, see the [IAM documentation](https://cloud.google.com/iam/docs/).
+ "auditConfigs": [ # Specifies cloud audit logging configuration for this policy.
+ { # Specifies the audit configuration for a service. The configuration determines which permission types are logged, and what identities, if any, are exempted from logging. An AuditConfig must have one or more AuditLogConfigs. If there are AuditConfigs for both `allServices` and a specific service, the union of the two AuditConfigs is used for that service: the log_types specified in each AuditConfig are enabled, and the exempted_members in each AuditLogConfig are exempted. Example Policy with multiple AuditConfigs: { "audit_configs": [ { "service": "allServices", "audit_log_configs": [ { "log_type": "DATA_READ", "exempted_members": [ "user:jose@example.com" ] }, { "log_type": "DATA_WRITE" }, { "log_type": "ADMIN_READ" } ] }, { "service": "sampleservice.googleapis.com", "audit_log_configs": [ { "log_type": "DATA_READ" }, { "log_type": "DATA_WRITE", "exempted_members": [ "user:aliya@example.com" ] } ] } ] } For sampleservice, this policy enables DATA_READ, DATA_WRITE and ADMIN_READ logging. It also exempts `jose@example.com` from DATA_READ logging, and `aliya@example.com` from DATA_WRITE logging.
+ "auditLogConfigs": [ # The configuration for logging of each type of permission.
+ { # Provides the configuration for logging a type of permissions. Example: { "audit_log_configs": [ { "log_type": "DATA_READ", "exempted_members": [ "user:jose@example.com" ] }, { "log_type": "DATA_WRITE" } ] } This enables 'DATA_READ' and 'DATA_WRITE' logging, while exempting jose@example.com from DATA_READ logging.
+ "exemptedMembers": [ # Specifies the identities that do not cause logging for this type of permission. Follows the same format of Binding.members.
+ "A String",
+ ],
+ "logType": "A String", # The log type that this config enables.
+ },
+ ],
+ "service": "A String", # Specifies a service that will be enabled for audit logging. For example, `storage.googleapis.com`, `cloudsql.googleapis.com`. `allServices` is a special value that covers all services.
+ },
+ ],
+ "bindings": [ # Associates a list of `members`, or principals, with a `role`. Optionally, may specify a `condition` that determines how and when the `bindings` are applied. Each of the `bindings` must contain at least one principal. The `bindings` in a `Policy` can refer to up to 1,500 principals; up to 250 of these principals can be Google groups. Each occurrence of a principal counts towards these limits. For example, if the `bindings` grant 50 different roles to `user:alice@example.com`, and not to any other principal, then you can add another 1,450 principals to the `bindings` in the `Policy`.
+ { # Associates `members`, or principals, with a `role`.
+ "condition": { # Represents a textual expression in the Common Expression Language (CEL) syntax. CEL is a C-like expression language. The syntax and semantics of CEL are documented at https://github.com/google/cel-spec. Example (Comparison): title: "Summary size limit" description: "Determines if a summary is less than 100 chars" expression: "document.summary.size() < 100" Example (Equality): title: "Requestor is owner" description: "Determines if requestor is the document owner" expression: "document.owner == request.auth.claims.email" Example (Logic): title: "Public documents" description: "Determine whether the document should be publicly visible" expression: "document.type != 'private' && document.type != 'internal'" Example (Data Manipulation): title: "Notification string" description: "Create a notification string with a timestamp." expression: "'New message received at ' + string(document.create_time)" The exact variables and functions that may be referenced within an expression are determined by the service that evaluates it. See the service documentation for additional information. # The condition that is associated with this binding. If the condition evaluates to `true`, then this binding applies to the current request. If the condition evaluates to `false`, then this binding does not apply to the current request. However, a different role binding might grant the same role to one or more of the principals in this binding. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+ "description": "A String", # Optional. Description of the expression. This is a longer text which describes the expression, e.g. when hovered over it in a UI.
+ "expression": "A String", # Textual representation of an expression in Common Expression Language syntax.
+ "location": "A String", # Optional. String indicating the location of the expression for error reporting, e.g. a file name and a position in the file.
+ "title": "A String", # Optional. Title for the expression, i.e. a short string describing its purpose. This can be used e.g. in UIs which allow to enter the expression.
+ },
+ "members": [ # Specifies the principals requesting access for a Google Cloud resource. `members` can have the following values: * `allUsers`: A special identifier that represents anyone who is on the internet; with or without a Google account. * `allAuthenticatedUsers`: A special identifier that represents anyone who is authenticated with a Google account or a service account. Does not include identities that come from external identity providers (IdPs) through identity federation. * `user:{emailid}`: An email address that represents a specific Google account. For example, `alice@example.com` . * `serviceAccount:{emailid}`: An email address that represents a Google service account. For example, `my-other-app@appspot.gserviceaccount.com`. * `serviceAccount:{projectid}.svc.id.goog[{namespace}/{kubernetes-sa}]`: An identifier for a [Kubernetes service account](https://cloud.google.com/kubernetes-engine/docs/how-to/kubernetes-service-accounts). For example, `my-project.svc.id.goog[my-namespace/my-kubernetes-sa]`. * `group:{emailid}`: An email address that represents a Google group. For example, `admins@example.com`. * `domain:{domain}`: The G Suite domain (primary) that represents all the users of that domain. For example, `google.com` or `example.com`. * `principal://iam.googleapis.com/locations/global/workforcePools/{pool_id}/subject/{subject_attribute_value}`: A single identity in a workforce identity pool. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/group/{group_id}`: All workforce identities in a group. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/attribute.{attribute_name}/{attribute_value}`: All workforce identities with a specific attribute value. * `principalSet://iam.googleapis.com/locations/global/workforcePools/{pool_id}/*`: All identities in a workforce identity pool. * `principal://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/subject/{subject_attribute_value}`: A single identity in a workload identity pool. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/group/{group_id}`: A workload identity pool group. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/attribute.{attribute_name}/{attribute_value}`: All identities in a workload identity pool with a certain attribute. * `principalSet://iam.googleapis.com/projects/{project_number}/locations/global/workloadIdentityPools/{pool_id}/*`: All identities in a workload identity pool. * `deleted:user:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a user that has been recently deleted. For example, `alice@example.com?uid=123456789012345678901`. If the user is recovered, this value reverts to `user:{emailid}` and the recovered user retains the role in the binding. * `deleted:serviceAccount:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a service account that has been recently deleted. For example, `my-other-app@appspot.gserviceaccount.com?uid=123456789012345678901`. If the service account is undeleted, this value reverts to `serviceAccount:{emailid}` and the undeleted service account retains the role in the binding. * `deleted:group:{emailid}?uid={uniqueid}`: An email address (plus unique identifier) representing a Google group that has been recently deleted. For example, `admins@example.com?uid=123456789012345678901`. If the group is recovered, this value reverts to `group:{emailid}` and the recovered group retains the role in the binding. * `deleted:principal://iam.googleapis.com/locations/global/workforcePools/{pool_id}/subject/{subject_attribute_value}`: Deleted single identity in a workforce identity pool. For example, `deleted:principal://iam.googleapis.com/locations/global/workforcePools/my-pool-id/subject/my-subject-attribute-value`.
+ "A String",
+ ],
+ "role": "A String", # Role that is assigned to the list of `members`, or principals. For example, `roles/viewer`, `roles/editor`, or `roles/owner`. For an overview of the IAM roles and permissions, see the [IAM documentation](https://cloud.google.com/iam/docs/roles-overview). For a list of the available pre-defined roles, see [here](https://cloud.google.com/iam/docs/understanding-roles).
+ },
+ ],
+ "etag": "A String", # `etag` is used for optimistic concurrency control as a way to help prevent simultaneous updates of a policy from overwriting each other. It is strongly suggested that systems make use of the `etag` in the read-modify-write cycle to perform policy updates in order to avoid race conditions: An `etag` is returned in the response to `getIamPolicy`, and systems are expected to put that etag in the request to `setIamPolicy` to ensure that their change will be applied to the same version of the policy. **Important:** If you use IAM Conditions, you must include the `etag` field whenever you call `setIamPolicy`. If you omit this field, then IAM allows you to overwrite a version `3` policy with a version `1` policy, and all of the conditions in the version `3` policy are lost.
+ "version": 42, # Specifies the format of the policy. Valid values are `0`, `1`, and `3`. Requests that specify an invalid value are rejected. Any operation that affects conditional role bindings must specify version `3`. This requirement applies to the following operations: * Getting a policy that includes a conditional role binding * Adding a conditional role binding to a policy * Changing a conditional role binding in a policy * Removing any role binding, with or without a condition, from a policy that includes conditions **Important:** If you use IAM Conditions, you must include the `etag` field whenever you call `setIamPolicy`. If you omit this field, then IAM allows you to overwrite a version `3` policy with a version `1` policy, and all of the conditions in the version `3` policy are lost. If a policy does not include any conditions, operations on that policy may specify any valid version or leave the field unset. To learn which resources support conditions in their IAM policies, see the [IAM documentation](https://cloud.google.com/iam/help/conditions/resource-policies).
+}
+
+
+
+ testIamPermissions(resource, body=None, x__xgafv=None)
+ Tests the IAM permissions for the specified view.
+
+Args:
+ resource: string, REQUIRED: The resource for which the policy detail is being requested. See [Resource names](https://cloud.google.com/apis/design/resource_names) for the appropriate value for this field. (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request message for `TestIamPermissions` method.
+ "permissions": [ # The set of permissions to check for the `resource`. Permissions with wildcards (such as `*` or `storage.*`) are not allowed. For more information see [IAM Overview](https://cloud.google.com/iam/docs/overview#permissions).
+ "A String",
+ ],
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Response message for `TestIamPermissions` method.
+ "permissions": [ # A subset of `TestPermissionsRequest.permissions` that the caller is allowed.
+ "A String",
+ ],
+}
+
+
+
\ No newline at end of file
diff --git a/docs/dyn/chromemanagement_v1.customers.chromeBrowsers.html b/docs/dyn/chromemanagement_v1.customers.chromeBrowsers.html
new file mode 100644
index 0000000000..3ba1559b4f
--- /dev/null
+++ b/docs/dyn/chromemanagement_v1.customers.chromeBrowsers.html
@@ -0,0 +1,826 @@
+
+
+
+Chrome Management API . customers . chromeBrowsers
+Instance Methods
+
+ close()
+Close httplib2 connections.
+
+Deletes the data collected from a Chrome browser profile.
+
+Retrieves a single Chrome Browser identified by its customer ID and resource ID.
+
+ list(parent, filter=None, orderBy=None, pageSize=None, pageToken=None, x__xgafv=None)
+Retrieves all Chrome Browsers of a customer (paginated).
+
+ list_next(previous_request, previous_response)
+Retrieves the next page of results.
+
+ move(parent, body=None, x__xgafv=None)
+Moves managed Chrome Browsers to a new Organizational Unit (OU). If there is an error while moving any of the browsers, none of the browsers will be moved.
+
+ patch(name, body=None, updateMask=None, x__xgafv=None)
+Updates annotation information for a Chrome Browser.
+Method Details
+
+ close()
+ Close httplib2 connections.
+
+
+
+ delete(name, x__xgafv=None)
+ Deletes the data collected from a Chrome browser profile.
+
+Args:
+ name: string, Required. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id} (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); }
+}
+
+
+
+ get(name, x__xgafv=None)
+ Retrieves a single Chrome Browser identified by its customer ID and resource ID.
+
+Args:
+ name: string, Required. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id} (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # A device with managed Chrome Browsers.
+ "annotatedAssetId": "A String", # Optional. Asset identifier as annotated by the administrator or specified during enrollment.
+ "annotatedLocation": "A String", # Optional. Address or location of the device as annotated by the administrator.
+ "annotatedNotes": "A String", # Optional. Notes about this device as annotated by the administrator.
+ "annotatedUser": "A String", # Optional. User of the device as annotated by the administrator.
+ "attestationCredential": { # Information of public key associated with a Chrome browser profile. # Output only. The attestation credential used for Device Trust Connector.
+ "keyRotationTime": "A String", # Output only. Latest rotation timestamp of the public key rotation.
+ "keyTrustLevel": "A String", # Output only. Trust level of the public key.
+ "keyType": "A String", # Output only. Type of the public key.
+ "publicKey": "A String", # Output only. Value of the public key.
+ },
+ "browserPermanentId": "A String", # Output only. Unique identifier of Chrome Browser.
+ "browserVersions": [ # Output only. List of all browser versions installed on this device.
+ "A String",
+ ],
+ "browsers": [ # Output only. List of Chrome browsers installed on this device.
+ { # A specific installation of Chrome on a device with managed Chrome browsers.
+ "browserVersion": "A String", # Output only. Browser version installed.
+ "channel": "A String", # Output only. The channel of the browser installed.
+ "executablePath": "A String", # Output only. The path to the Chrome.exe executable for this browser.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded on this Chrome browser.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this Chrome browser.
+ "pendingInstallVersion": "A String", # Output only. Pending version of a browser is installed. This field is only set when the current active browser has a different version as in "browser_version".
+ "plugins": [ # Output only. List of all plugins installed on this Chrome browser.
+ { # Information about a Chrome browser plugin installed on a reported Chrome installation on a device with managed Chrome browsers.
+ "description": "A String", # Output only. The description of the installed plugin.
+ "filename": "A String", # Output only. The filename of the installed plugin.
+ "pluginName": "A String", # Output only. The name of the installed plugin.
+ },
+ ],
+ "profiles": [ # Output only. The list of profiles for this browser.
+ { # A profile reported on an installation of Chrome on a device with managed chrome browsers.
+ "chromeSignedInUserEmail": "A String", # Output only. The email of the user signed in to this profile, if any.
+ "extensionPolicies": [ # Output only. The list of extensions with applied extension policies.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "extensions": [ # Output only. The list of extensions installed for this profile.
+ { # Information about extensions installed on profile reported on an installed Chrome on a device with managed chrome browsers.
+ "appType": "A String", # Output only. The type of the extension.
+ "configuredAppPolicy": "A String", # Output only. If the extension is controlled by a browser policy.
+ "description": "A String", # Output only. The localized description of this extension.
+ "disabled": True or False, # Output only. Whether the extension is disabled.
+ "extensionId": "A String", # Output only. The id of the extension.
+ "extensionName": "A String", # Output only. The localized name of this extension as reported by Chrome.
+ "homepageUri": "A String", # Output only. The URL of the homepage for this extension.
+ "icons": [ # Output only. The list of icons available for this extension.
+ { # An icon of a Chrome browser extension.
+ "size": 42, # Output only. The size of the extension icon.
+ "uri": "A String", # Output only. The URL to the extension icon.
+ },
+ ],
+ "installType": "A String", # Output only. The way the extension was installed.
+ "manifestVersion": 42, # Output only. The manifest version of this extension.
+ "permissions": [ # Output only. The list of permissions required for this extension.
+ "A String",
+ ],
+ "version": "A String", # Output only. The version of the extension.
+ },
+ ],
+ "id": "A String", # Output only. A unique ID for this profile.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded for this profile.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this profile.
+ "profileName": "A String", # Output only. The name given to this profile.
+ "safeBrowsingWarnings": "A String", # Output only. The number of safe browsing warnings that were shown on this profile.
+ "safeBrowsingWarningsClickThroughs": "A String", # Output only. The number of safe browsing warnings that were clicked through on this profile.
+ "safeBrowsingWarningsResetTime": "A String", # Output only. The last time the safe browsing warning info was reset to 0.
+ "userPolicies": [ # Output only. The list of Chrome browser policies applied to this profile.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ },
+ ],
+ "deviceIdentifiersHistory": { # History of device identifiers reported from a device. # Output only. The history of identifiers reported to this device.
+ "deviceIdCollision": True or False, # Output only. The device id is potentially being used by multiple devices.
+ "lastDeviceIdCollisionDetectionTime": "A String", # Output only. The last time we detected the device id collision.
+ "records": [ # Output only. List of device identifiers sent by this device. In descending order by last_activity_time.
+ { # Record of the device identifiers and when they were reported.
+ "firstRecordTime": "A String", # Output only. The first time these identifiers were reported.
+ "identifiers": { # Collection of device identifiers. # Output only. Device identifiers.
+ "hostName": "A String", # Output only. The device's host name.
+ "machineName": "A String", # Output only. The name of the machine associated with the device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ },
+ "lastActivityTime": "A String", # Output only. The time of the last activity of these identifiers either from reporting, policy fetching or registration.
+ },
+ ],
+ },
+ "extensionCount": "A String", # Output only. The number of distinct extensions installed on all the Chrome browsers of this device.
+ "lastActivityTime": "A String", # Output only. The time of the last activity on the device device either from reporting, policy fetching or registration.
+ "lastDeviceUser": "A String", # Output only. The last user who logged into a Chrome browser on this device.
+ "lastDeviceUsers": [ # Output only. List of recent browser users, in descending order by last report time.
+ { # A Chrome browser user.
+ "lastStatusReportTime": "A String", # Output only. The time when this user info last reported a status.
+ "userName": "A String", # Output only. The user name of the user that signed into chrome.
+ },
+ ],
+ "lastPolicyFetchTime": "A String", # Output only. The time the last policy fetch succeeded on this device.
+ "lastRegistrationTime": "A String", # Output only. The time the device last tried to register.
+ "lastStatusReportTime": "A String", # Output only. The time the device last sent a status report.
+ "machineExtensionPolicies": [ # Output only. The list of all device level policies set for extensions installed in Chrome on this device.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "machineName": "A String", # Output only. Machine name.
+ "machinePolicies": [ # Output only. The list of all device level policies set for Chrome browsers on this device.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ "name": "A String", # Identifier. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id}
+ "omahaRecentFetchTime": "A String", # Output only. Recent policy fetch activity by Omaha. Not necessarily the last fetch since this field is only updated daily.
+ "orgUnitId": "A String", # Output only. The obfuscated organizational unit ID of the browser.
+ "osArchitecture": "A String", # Output only. Device CPU architecture.
+ "osPlatform": "A String", # Output only. Device Os platform.
+ "osPlatformVersion": "A String", # Output only. Device Os platform and version combined.
+ "osVersion": "A String", # Output only. Device Os Version.
+ "policyCount": "A String", # Output only. The number of distinct policies set on all the Chrome browsers of this device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ "virtualDeviceId": "A String", # Output only. The initial device id sent by the first Chrome browser that enrolled this device.
+}
+
+
+
+ list(parent, filter=None, orderBy=None, pageSize=None, pageToken=None, x__xgafv=None)
+ Retrieves all Chrome Browsers of a customer (paginated).
+
+Args:
+ parent: string, Required. Format: customers/{customer_id} (required)
+ filter: string, Optional. The filter used to filter browsers. The following fields can be used in the filter: * `browser_permanent_id` * `last_policy_fetch_time` * `os_platform` * `os_architecture` * `os_version` * `machine_name` * `annotated_location` * `annotated_user` * `annotated_asset_id` * `annotated_note` * `org_unit_path` * `org_unit_id` * `last_registration_time` * `os_platform_version` * `browser_version` * `last_status_report_time` * `extension_count` * `policy_count` * `last_device_user` * `last_activity_time` * `device_id_collision` The following functions can be used in the filter: * `in_group(string)`: Filters browsers that belong to the specified Cloud Identity group resource name (e.g. `in_group("groups/{group_id}")`). Any of the above fields or functions can be used to specify a filter, and filtering by multiple fields or functions is supported with AND operator. String type, Integer type fields and enum type fields support `=` and `:` operators. The timestamp type fields support `=`, `<=` and `>=` operators. Timestamps expect an RFC-3339 formatted string (e.g. 2012-04-21T11:30:00-04:00). Wildcard `*` is only supported for `machine_name`, `annotated_asset_id`, and `browser_version`. In addition, global string literal filtering without a field name is supported: a single term (e.g., `ABC`) matches if any indexed string field contains `ABC`, and multiple terms joined by `AND` or whitespace (e.g., `machine AND 73` or `machine 73`) match browsers where every term appears in at least one indexed string field (whereas a single quoted phrase like `"machine 73"` matches the contiguous phrase within a single field).
+ orderBy: string, Optional. The fields used to specify the ordering of the results. The supported fields are: * `browser_permanent_id` * `last_sync` * `annotated_user` * `annotated_location` * `annotated_asset_id` * `annotated_notes` * `org_unit_path` * `os_version` * `enrollment_date` * `extension_count` * `policy_count` * `last_signed_in_user` * `machine_name` * `browser_version_channel` * `os_platform_version` * `last_activity_time` * `browser_version` By default, sorting is in ascending order, to specify descending order for a field, a suffix ` desc` should be added to the field name. The default ordering is the descending order of `last_status_report_time`.
+ pageSize: integer, Optional. Maximum number of results to return. Maximum and default are 100.
+ pageToken: string, Optional. Token to specify next page in the list.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Response to ListChromeBrowsers method.
+ "chromeBrowsers": [ # List of Chrome Browser device objects.
+ { # A device with managed Chrome Browsers.
+ "annotatedAssetId": "A String", # Optional. Asset identifier as annotated by the administrator or specified during enrollment.
+ "annotatedLocation": "A String", # Optional. Address or location of the device as annotated by the administrator.
+ "annotatedNotes": "A String", # Optional. Notes about this device as annotated by the administrator.
+ "annotatedUser": "A String", # Optional. User of the device as annotated by the administrator.
+ "attestationCredential": { # Information of public key associated with a Chrome browser profile. # Output only. The attestation credential used for Device Trust Connector.
+ "keyRotationTime": "A String", # Output only. Latest rotation timestamp of the public key rotation.
+ "keyTrustLevel": "A String", # Output only. Trust level of the public key.
+ "keyType": "A String", # Output only. Type of the public key.
+ "publicKey": "A String", # Output only. Value of the public key.
+ },
+ "browserPermanentId": "A String", # Output only. Unique identifier of Chrome Browser.
+ "browserVersions": [ # Output only. List of all browser versions installed on this device.
+ "A String",
+ ],
+ "browsers": [ # Output only. List of Chrome browsers installed on this device.
+ { # A specific installation of Chrome on a device with managed Chrome browsers.
+ "browserVersion": "A String", # Output only. Browser version installed.
+ "channel": "A String", # Output only. The channel of the browser installed.
+ "executablePath": "A String", # Output only. The path to the Chrome.exe executable for this browser.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded on this Chrome browser.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this Chrome browser.
+ "pendingInstallVersion": "A String", # Output only. Pending version of a browser is installed. This field is only set when the current active browser has a different version as in "browser_version".
+ "plugins": [ # Output only. List of all plugins installed on this Chrome browser.
+ { # Information about a Chrome browser plugin installed on a reported Chrome installation on a device with managed Chrome browsers.
+ "description": "A String", # Output only. The description of the installed plugin.
+ "filename": "A String", # Output only. The filename of the installed plugin.
+ "pluginName": "A String", # Output only. The name of the installed plugin.
+ },
+ ],
+ "profiles": [ # Output only. The list of profiles for this browser.
+ { # A profile reported on an installation of Chrome on a device with managed chrome browsers.
+ "chromeSignedInUserEmail": "A String", # Output only. The email of the user signed in to this profile, if any.
+ "extensionPolicies": [ # Output only. The list of extensions with applied extension policies.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "extensions": [ # Output only. The list of extensions installed for this profile.
+ { # Information about extensions installed on profile reported on an installed Chrome on a device with managed chrome browsers.
+ "appType": "A String", # Output only. The type of the extension.
+ "configuredAppPolicy": "A String", # Output only. If the extension is controlled by a browser policy.
+ "description": "A String", # Output only. The localized description of this extension.
+ "disabled": True or False, # Output only. Whether the extension is disabled.
+ "extensionId": "A String", # Output only. The id of the extension.
+ "extensionName": "A String", # Output only. The localized name of this extension as reported by Chrome.
+ "homepageUri": "A String", # Output only. The URL of the homepage for this extension.
+ "icons": [ # Output only. The list of icons available for this extension.
+ { # An icon of a Chrome browser extension.
+ "size": 42, # Output only. The size of the extension icon.
+ "uri": "A String", # Output only. The URL to the extension icon.
+ },
+ ],
+ "installType": "A String", # Output only. The way the extension was installed.
+ "manifestVersion": 42, # Output only. The manifest version of this extension.
+ "permissions": [ # Output only. The list of permissions required for this extension.
+ "A String",
+ ],
+ "version": "A String", # Output only. The version of the extension.
+ },
+ ],
+ "id": "A String", # Output only. A unique ID for this profile.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded for this profile.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this profile.
+ "profileName": "A String", # Output only. The name given to this profile.
+ "safeBrowsingWarnings": "A String", # Output only. The number of safe browsing warnings that were shown on this profile.
+ "safeBrowsingWarningsClickThroughs": "A String", # Output only. The number of safe browsing warnings that were clicked through on this profile.
+ "safeBrowsingWarningsResetTime": "A String", # Output only. The last time the safe browsing warning info was reset to 0.
+ "userPolicies": [ # Output only. The list of Chrome browser policies applied to this profile.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ },
+ ],
+ "deviceIdentifiersHistory": { # History of device identifiers reported from a device. # Output only. The history of identifiers reported to this device.
+ "deviceIdCollision": True or False, # Output only. The device id is potentially being used by multiple devices.
+ "lastDeviceIdCollisionDetectionTime": "A String", # Output only. The last time we detected the device id collision.
+ "records": [ # Output only. List of device identifiers sent by this device. In descending order by last_activity_time.
+ { # Record of the device identifiers and when they were reported.
+ "firstRecordTime": "A String", # Output only. The first time these identifiers were reported.
+ "identifiers": { # Collection of device identifiers. # Output only. Device identifiers.
+ "hostName": "A String", # Output only. The device's host name.
+ "machineName": "A String", # Output only. The name of the machine associated with the device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ },
+ "lastActivityTime": "A String", # Output only. The time of the last activity of these identifiers either from reporting, policy fetching or registration.
+ },
+ ],
+ },
+ "extensionCount": "A String", # Output only. The number of distinct extensions installed on all the Chrome browsers of this device.
+ "lastActivityTime": "A String", # Output only. The time of the last activity on the device device either from reporting, policy fetching or registration.
+ "lastDeviceUser": "A String", # Output only. The last user who logged into a Chrome browser on this device.
+ "lastDeviceUsers": [ # Output only. List of recent browser users, in descending order by last report time.
+ { # A Chrome browser user.
+ "lastStatusReportTime": "A String", # Output only. The time when this user info last reported a status.
+ "userName": "A String", # Output only. The user name of the user that signed into chrome.
+ },
+ ],
+ "lastPolicyFetchTime": "A String", # Output only. The time the last policy fetch succeeded on this device.
+ "lastRegistrationTime": "A String", # Output only. The time the device last tried to register.
+ "lastStatusReportTime": "A String", # Output only. The time the device last sent a status report.
+ "machineExtensionPolicies": [ # Output only. The list of all device level policies set for extensions installed in Chrome on this device.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "machineName": "A String", # Output only. Machine name.
+ "machinePolicies": [ # Output only. The list of all device level policies set for Chrome browsers on this device.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ "name": "A String", # Identifier. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id}
+ "omahaRecentFetchTime": "A String", # Output only. Recent policy fetch activity by Omaha. Not necessarily the last fetch since this field is only updated daily.
+ "orgUnitId": "A String", # Output only. The obfuscated organizational unit ID of the browser.
+ "osArchitecture": "A String", # Output only. Device CPU architecture.
+ "osPlatform": "A String", # Output only. Device Os platform.
+ "osPlatformVersion": "A String", # Output only. Device Os platform and version combined.
+ "osVersion": "A String", # Output only. Device Os Version.
+ "policyCount": "A String", # Output only. The number of distinct policies set on all the Chrome browsers of this device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ "virtualDeviceId": "A String", # Output only. The initial device id sent by the first Chrome browser that enrolled this device.
+ },
+ ],
+ "nextPageToken": "A String", # Token used to access next page of this result.
+}
+
+
+
+ list_next(previous_request, previous_response)
+ Retrieves the next page of results.
+
+Args:
+ previous_request: The request for the previous page. (required)
+ previous_response: The response from the request for the previous page. (required)
+
+Returns:
+ A request object that you can call 'execute()' on to request the next
+ page. Returns None if there are no more items in the collection.
+
+
+
+
+ move(parent, body=None, x__xgafv=None)
+ Moves managed Chrome Browsers to a new Organizational Unit (OU). If there is an error while moving any of the browsers, none of the browsers will be moved.
+
+Args:
+ parent: string, Required. Format: customers/{customer_id} (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request to MoveChromeBrowsers method.
+ "destinationOrgUnitId": "A String", # Required. ID of the destination organizational unit.
+ "resourceIds": [ # Required. List of resource IDs of Chrome Browser Devices to move. A maximum of 600 browsers may be moved per request.
+ "A String",
+ ],
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Response to MoveChromeBrowsers method.
+}
+
+
+
+ patch(name, body=None, updateMask=None, x__xgafv=None)
+ Updates annotation information for a Chrome Browser.
+
+Args:
+ name: string, Identifier. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id} (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # A device with managed Chrome Browsers.
+ "annotatedAssetId": "A String", # Optional. Asset identifier as annotated by the administrator or specified during enrollment.
+ "annotatedLocation": "A String", # Optional. Address or location of the device as annotated by the administrator.
+ "annotatedNotes": "A String", # Optional. Notes about this device as annotated by the administrator.
+ "annotatedUser": "A String", # Optional. User of the device as annotated by the administrator.
+ "attestationCredential": { # Information of public key associated with a Chrome browser profile. # Output only. The attestation credential used for Device Trust Connector.
+ "keyRotationTime": "A String", # Output only. Latest rotation timestamp of the public key rotation.
+ "keyTrustLevel": "A String", # Output only. Trust level of the public key.
+ "keyType": "A String", # Output only. Type of the public key.
+ "publicKey": "A String", # Output only. Value of the public key.
+ },
+ "browserPermanentId": "A String", # Output only. Unique identifier of Chrome Browser.
+ "browserVersions": [ # Output only. List of all browser versions installed on this device.
+ "A String",
+ ],
+ "browsers": [ # Output only. List of Chrome browsers installed on this device.
+ { # A specific installation of Chrome on a device with managed Chrome browsers.
+ "browserVersion": "A String", # Output only. Browser version installed.
+ "channel": "A String", # Output only. The channel of the browser installed.
+ "executablePath": "A String", # Output only. The path to the Chrome.exe executable for this browser.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded on this Chrome browser.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this Chrome browser.
+ "pendingInstallVersion": "A String", # Output only. Pending version of a browser is installed. This field is only set when the current active browser has a different version as in "browser_version".
+ "plugins": [ # Output only. List of all plugins installed on this Chrome browser.
+ { # Information about a Chrome browser plugin installed on a reported Chrome installation on a device with managed Chrome browsers.
+ "description": "A String", # Output only. The description of the installed plugin.
+ "filename": "A String", # Output only. The filename of the installed plugin.
+ "pluginName": "A String", # Output only. The name of the installed plugin.
+ },
+ ],
+ "profiles": [ # Output only. The list of profiles for this browser.
+ { # A profile reported on an installation of Chrome on a device with managed chrome browsers.
+ "chromeSignedInUserEmail": "A String", # Output only. The email of the user signed in to this profile, if any.
+ "extensionPolicies": [ # Output only. The list of extensions with applied extension policies.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "extensions": [ # Output only. The list of extensions installed for this profile.
+ { # Information about extensions installed on profile reported on an installed Chrome on a device with managed chrome browsers.
+ "appType": "A String", # Output only. The type of the extension.
+ "configuredAppPolicy": "A String", # Output only. If the extension is controlled by a browser policy.
+ "description": "A String", # Output only. The localized description of this extension.
+ "disabled": True or False, # Output only. Whether the extension is disabled.
+ "extensionId": "A String", # Output only. The id of the extension.
+ "extensionName": "A String", # Output only. The localized name of this extension as reported by Chrome.
+ "homepageUri": "A String", # Output only. The URL of the homepage for this extension.
+ "icons": [ # Output only. The list of icons available for this extension.
+ { # An icon of a Chrome browser extension.
+ "size": 42, # Output only. The size of the extension icon.
+ "uri": "A String", # Output only. The URL to the extension icon.
+ },
+ ],
+ "installType": "A String", # Output only. The way the extension was installed.
+ "manifestVersion": 42, # Output only. The manifest version of this extension.
+ "permissions": [ # Output only. The list of permissions required for this extension.
+ "A String",
+ ],
+ "version": "A String", # Output only. The version of the extension.
+ },
+ ],
+ "id": "A String", # Output only. A unique ID for this profile.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded for this profile.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this profile.
+ "profileName": "A String", # Output only. The name given to this profile.
+ "safeBrowsingWarnings": "A String", # Output only. The number of safe browsing warnings that were shown on this profile.
+ "safeBrowsingWarningsClickThroughs": "A String", # Output only. The number of safe browsing warnings that were clicked through on this profile.
+ "safeBrowsingWarningsResetTime": "A String", # Output only. The last time the safe browsing warning info was reset to 0.
+ "userPolicies": [ # Output only. The list of Chrome browser policies applied to this profile.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ },
+ ],
+ "deviceIdentifiersHistory": { # History of device identifiers reported from a device. # Output only. The history of identifiers reported to this device.
+ "deviceIdCollision": True or False, # Output only. The device id is potentially being used by multiple devices.
+ "lastDeviceIdCollisionDetectionTime": "A String", # Output only. The last time we detected the device id collision.
+ "records": [ # Output only. List of device identifiers sent by this device. In descending order by last_activity_time.
+ { # Record of the device identifiers and when they were reported.
+ "firstRecordTime": "A String", # Output only. The first time these identifiers were reported.
+ "identifiers": { # Collection of device identifiers. # Output only. Device identifiers.
+ "hostName": "A String", # Output only. The device's host name.
+ "machineName": "A String", # Output only. The name of the machine associated with the device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ },
+ "lastActivityTime": "A String", # Output only. The time of the last activity of these identifiers either from reporting, policy fetching or registration.
+ },
+ ],
+ },
+ "extensionCount": "A String", # Output only. The number of distinct extensions installed on all the Chrome browsers of this device.
+ "lastActivityTime": "A String", # Output only. The time of the last activity on the device device either from reporting, policy fetching or registration.
+ "lastDeviceUser": "A String", # Output only. The last user who logged into a Chrome browser on this device.
+ "lastDeviceUsers": [ # Output only. List of recent browser users, in descending order by last report time.
+ { # A Chrome browser user.
+ "lastStatusReportTime": "A String", # Output only. The time when this user info last reported a status.
+ "userName": "A String", # Output only. The user name of the user that signed into chrome.
+ },
+ ],
+ "lastPolicyFetchTime": "A String", # Output only. The time the last policy fetch succeeded on this device.
+ "lastRegistrationTime": "A String", # Output only. The time the device last tried to register.
+ "lastStatusReportTime": "A String", # Output only. The time the device last sent a status report.
+ "machineExtensionPolicies": [ # Output only. The list of all device level policies set for extensions installed in Chrome on this device.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "machineName": "A String", # Output only. Machine name.
+ "machinePolicies": [ # Output only. The list of all device level policies set for Chrome browsers on this device.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ "name": "A String", # Identifier. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id}
+ "omahaRecentFetchTime": "A String", # Output only. Recent policy fetch activity by Omaha. Not necessarily the last fetch since this field is only updated daily.
+ "orgUnitId": "A String", # Output only. The obfuscated organizational unit ID of the browser.
+ "osArchitecture": "A String", # Output only. Device CPU architecture.
+ "osPlatform": "A String", # Output only. Device Os platform.
+ "osPlatformVersion": "A String", # Output only. Device Os platform and version combined.
+ "osVersion": "A String", # Output only. Device Os Version.
+ "policyCount": "A String", # Output only. The number of distinct policies set on all the Chrome browsers of this device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ "virtualDeviceId": "A String", # Output only. The initial device id sent by the first Chrome browser that enrolled this device.
+}
+
+ updateMask: string, Optional. The update mask that can be used to specify which fields to update.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # A device with managed Chrome Browsers.
+ "annotatedAssetId": "A String", # Optional. Asset identifier as annotated by the administrator or specified during enrollment.
+ "annotatedLocation": "A String", # Optional. Address or location of the device as annotated by the administrator.
+ "annotatedNotes": "A String", # Optional. Notes about this device as annotated by the administrator.
+ "annotatedUser": "A String", # Optional. User of the device as annotated by the administrator.
+ "attestationCredential": { # Information of public key associated with a Chrome browser profile. # Output only. The attestation credential used for Device Trust Connector.
+ "keyRotationTime": "A String", # Output only. Latest rotation timestamp of the public key rotation.
+ "keyTrustLevel": "A String", # Output only. Trust level of the public key.
+ "keyType": "A String", # Output only. Type of the public key.
+ "publicKey": "A String", # Output only. Value of the public key.
+ },
+ "browserPermanentId": "A String", # Output only. Unique identifier of Chrome Browser.
+ "browserVersions": [ # Output only. List of all browser versions installed on this device.
+ "A String",
+ ],
+ "browsers": [ # Output only. List of Chrome browsers installed on this device.
+ { # A specific installation of Chrome on a device with managed Chrome browsers.
+ "browserVersion": "A String", # Output only. Browser version installed.
+ "channel": "A String", # Output only. The channel of the browser installed.
+ "executablePath": "A String", # Output only. The path to the Chrome.exe executable for this browser.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded on this Chrome browser.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this Chrome browser.
+ "pendingInstallVersion": "A String", # Output only. Pending version of a browser is installed. This field is only set when the current active browser has a different version as in "browser_version".
+ "plugins": [ # Output only. List of all plugins installed on this Chrome browser.
+ { # Information about a Chrome browser plugin installed on a reported Chrome installation on a device with managed Chrome browsers.
+ "description": "A String", # Output only. The description of the installed plugin.
+ "filename": "A String", # Output only. The filename of the installed plugin.
+ "pluginName": "A String", # Output only. The name of the installed plugin.
+ },
+ ],
+ "profiles": [ # Output only. The list of profiles for this browser.
+ { # A profile reported on an installation of Chrome on a device with managed chrome browsers.
+ "chromeSignedInUserEmail": "A String", # Output only. The email of the user signed in to this profile, if any.
+ "extensionPolicies": [ # Output only. The list of extensions with applied extension policies.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "extensions": [ # Output only. The list of extensions installed for this profile.
+ { # Information about extensions installed on profile reported on an installed Chrome on a device with managed chrome browsers.
+ "appType": "A String", # Output only. The type of the extension.
+ "configuredAppPolicy": "A String", # Output only. If the extension is controlled by a browser policy.
+ "description": "A String", # Output only. The localized description of this extension.
+ "disabled": True or False, # Output only. Whether the extension is disabled.
+ "extensionId": "A String", # Output only. The id of the extension.
+ "extensionName": "A String", # Output only. The localized name of this extension as reported by Chrome.
+ "homepageUri": "A String", # Output only. The URL of the homepage for this extension.
+ "icons": [ # Output only. The list of icons available for this extension.
+ { # An icon of a Chrome browser extension.
+ "size": 42, # Output only. The size of the extension icon.
+ "uri": "A String", # Output only. The URL to the extension icon.
+ },
+ ],
+ "installType": "A String", # Output only. The way the extension was installed.
+ "manifestVersion": 42, # Output only. The manifest version of this extension.
+ "permissions": [ # Output only. The list of permissions required for this extension.
+ "A String",
+ ],
+ "version": "A String", # Output only. The version of the extension.
+ },
+ ],
+ "id": "A String", # Output only. A unique ID for this profile.
+ "lastPolicyFetchTime": "A String", # Output only. The time when the last request to fetch policies succeeded for this profile.
+ "lastStatusReportTime": "A String", # Output only. The time when the last report was received from this profile.
+ "profileName": "A String", # Output only. The name given to this profile.
+ "safeBrowsingWarnings": "A String", # Output only. The number of safe browsing warnings that were shown on this profile.
+ "safeBrowsingWarningsClickThroughs": "A String", # Output only. The number of safe browsing warnings that were clicked through on this profile.
+ "safeBrowsingWarningsResetTime": "A String", # Output only. The last time the safe browsing warning info was reset to 0.
+ "userPolicies": [ # Output only. The list of Chrome browser policies applied to this profile.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ },
+ ],
+ "deviceIdentifiersHistory": { # History of device identifiers reported from a device. # Output only. The history of identifiers reported to this device.
+ "deviceIdCollision": True or False, # Output only. The device id is potentially being used by multiple devices.
+ "lastDeviceIdCollisionDetectionTime": "A String", # Output only. The last time we detected the device id collision.
+ "records": [ # Output only. List of device identifiers sent by this device. In descending order by last_activity_time.
+ { # Record of the device identifiers and when they were reported.
+ "firstRecordTime": "A String", # Output only. The first time these identifiers were reported.
+ "identifiers": { # Collection of device identifiers. # Output only. Device identifiers.
+ "hostName": "A String", # Output only. The device's host name.
+ "machineName": "A String", # Output only. The name of the machine associated with the device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ },
+ "lastActivityTime": "A String", # Output only. The time of the last activity of these identifiers either from reporting, policy fetching or registration.
+ },
+ ],
+ },
+ "extensionCount": "A String", # Output only. The number of distinct extensions installed on all the Chrome browsers of this device.
+ "lastActivityTime": "A String", # Output only. The time of the last activity on the device device either from reporting, policy fetching or registration.
+ "lastDeviceUser": "A String", # Output only. The last user who logged into a Chrome browser on this device.
+ "lastDeviceUsers": [ # Output only. List of recent browser users, in descending order by last report time.
+ { # A Chrome browser user.
+ "lastStatusReportTime": "A String", # Output only. The time when this user info last reported a status.
+ "userName": "A String", # Output only. The user name of the user that signed into chrome.
+ },
+ ],
+ "lastPolicyFetchTime": "A String", # Output only. The time the last policy fetch succeeded on this device.
+ "lastRegistrationTime": "A String", # Output only. The time the device last tried to register.
+ "lastStatusReportTime": "A String", # Output only. The time the device last sent a status report.
+ "machineExtensionPolicies": [ # Output only. The list of all device level policies set for extensions installed in Chrome on this device.
+ { # The reported policies applied to an extension installed on a Chrome profile.
+ "extensionId": "A String", # Output only. The id of the extension the policies are applied to.
+ "extensionName": "A String", # Output only. The name of the extension the policies are applied to.
+ "policies": [ # Output only. List of policies applied to this extension.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ },
+ ],
+ "machineName": "A String", # Output only. Machine name.
+ "machinePolicies": [ # Output only. The list of all device level policies set for Chrome browsers on this device.
+ { # The applied policy reported by a managed Chrome browser. The policy can be applied at various scopes (machine, profile, extension).
+ "error": "A String", # Output only. A localized string with all the errors reported when trying to apply this policy, if any.
+ "policyName": "A String", # Output only. The name of the policy.
+ "source": "A String", # Output only. The source of the policy.
+ "value": "A String", # Output only. The value of the policy.
+ },
+ ],
+ "name": "A String", # Identifier. Format: customers/{customer_id}/chromeBrowsers/{browser_permanent_id}
+ "omahaRecentFetchTime": "A String", # Output only. Recent policy fetch activity by Omaha. Not necessarily the last fetch since this field is only updated daily.
+ "orgUnitId": "A String", # Output only. The obfuscated organizational unit ID of the browser.
+ "osArchitecture": "A String", # Output only. Device CPU architecture.
+ "osPlatform": "A String", # Output only. Device Os platform.
+ "osPlatformVersion": "A String", # Output only. Device Os platform and version combined.
+ "osVersion": "A String", # Output only. Device Os Version.
+ "policyCount": "A String", # Output only. The number of distinct policies set on all the Chrome browsers of this device.
+ "serialNumber": "A String", # Output only. The device's serial number.
+ "virtualDeviceId": "A String", # Output only. The initial device id sent by the first Chrome browser that enrolled this device.
+}
+
+
+
\ No newline at end of file
diff --git a/docs/dyn/chromemanagement_v1.customers.enrollmentTokens.html b/docs/dyn/chromemanagement_v1.customers.enrollmentTokens.html
new file mode 100644
index 0000000000..83e4c22d83
--- /dev/null
+++ b/docs/dyn/chromemanagement_v1.customers.enrollmentTokens.html
@@ -0,0 +1,263 @@
+
+
+
+Chrome Management API . customers . enrollmentTokens
+Instance Methods
+
+ close()
+Close httplib2 connections.
+
+ create(parent, body=None, enrollmentTokenId=None, x__xgafv=None)
+Creates a new enrollment token for a browser device. Creation fails if there is already an active token for this customer under the same org unit.
+
+Gets a browser device enrollment token.
+
+ list(parent, filter=None, pageSize=None, pageToken=None, x__xgafv=None)
+Lists all browser device enrollment tokens.
+
+ list_next(previous_request, previous_response)
+Retrieves the next page of results.
+
+ revoke(name, body=None, x__xgafv=None)
+Revokes a browser device enrollment token.
+Method Details
+
+ close()
+ Close httplib2 connections.
+
+
+
+ create(parent, body=None, enrollmentTokenId=None, x__xgafv=None)
+ Creates a new enrollment token for a browser device. Creation fails if there is already an active token for this customer under the same org unit.
+
+Args:
+ parent: string, Required. The parent resource where this enrollment token will be created. Format: customers/{customer} (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Enrollment token for a chrome device.
+ "createTime": "A String", # Output only. The creation time of the token.
+ "creatorId": "A String", # Output only. The id of the user who created the token.
+ "customerId": "A String", # Output only. The obfuscated ID of the customer to whom this token is associated.
+ "name": "A String", # Identifier. The resource name of the enrollment token. Format: customers/{customer}/enrollmentTokens/{token_permanent_id}
+ "orgUnitId": "A String", # Optional. Obfuscated ID of the organization unit this token is associated with. If not set, the token is created for the root organization unit.
+ "revokeTime": "A String", # Output only. The revocation time of the token, null if the token is not revoked.
+ "revokerId": "A String", # Output only. The id of the user who revoked the token, null if the token is not revoked.
+ "state": "A String", # Output only. The state of the token.
+ "token": "A String", # Output only. The value of the token stored on the device and used to identify it to the management service handling the enrollment request.
+ "tokenPermanentId": "A String", # Output only. Unique identifier of the enrollment token used to access information on tokens.
+ "tokenType": "A String", # Optional. The device type this token is used for. If not set, defaults to `ENROLLMENT_TOKEN_TYPE_CHROME_BROWSER`.
+}
+
+ enrollmentTokenId: string, Optional. The ID to use for the enrollment token, which will become the final component of the enrollment token's resource name. This value must be local-unique under the customer and is optional. If not provided, it will be auto-generated. If provided, it must be 1-63 characters long and match the regular expression `[a-zA-Z0-9._-]+`.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Enrollment token for a chrome device.
+ "createTime": "A String", # Output only. The creation time of the token.
+ "creatorId": "A String", # Output only. The id of the user who created the token.
+ "customerId": "A String", # Output only. The obfuscated ID of the customer to whom this token is associated.
+ "name": "A String", # Identifier. The resource name of the enrollment token. Format: customers/{customer}/enrollmentTokens/{token_permanent_id}
+ "orgUnitId": "A String", # Optional. Obfuscated ID of the organization unit this token is associated with. If not set, the token is created for the root organization unit.
+ "revokeTime": "A String", # Output only. The revocation time of the token, null if the token is not revoked.
+ "revokerId": "A String", # Output only. The id of the user who revoked the token, null if the token is not revoked.
+ "state": "A String", # Output only. The state of the token.
+ "token": "A String", # Output only. The value of the token stored on the device and used to identify it to the management service handling the enrollment request.
+ "tokenPermanentId": "A String", # Output only. Unique identifier of the enrollment token used to access information on tokens.
+ "tokenType": "A String", # Optional. The device type this token is used for. If not set, defaults to `ENROLLMENT_TOKEN_TYPE_CHROME_BROWSER`.
+}
+
+
+
+ get(name, x__xgafv=None)
+ Gets a browser device enrollment token.
+
+Args:
+ name: string, Required. The name of the enrollment token to retrieve. Format: customers/{customer}/enrollmentTokens/{token_permanent_id} (required)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Enrollment token for a chrome device.
+ "createTime": "A String", # Output only. The creation time of the token.
+ "creatorId": "A String", # Output only. The id of the user who created the token.
+ "customerId": "A String", # Output only. The obfuscated ID of the customer to whom this token is associated.
+ "name": "A String", # Identifier. The resource name of the enrollment token. Format: customers/{customer}/enrollmentTokens/{token_permanent_id}
+ "orgUnitId": "A String", # Optional. Obfuscated ID of the organization unit this token is associated with. If not set, the token is created for the root organization unit.
+ "revokeTime": "A String", # Output only. The revocation time of the token, null if the token is not revoked.
+ "revokerId": "A String", # Output only. The id of the user who revoked the token, null if the token is not revoked.
+ "state": "A String", # Output only. The state of the token.
+ "token": "A String", # Output only. The value of the token stored on the device and used to identify it to the management service handling the enrollment request.
+ "tokenPermanentId": "A String", # Output only. Unique identifier of the enrollment token used to access information on tokens.
+ "tokenType": "A String", # Optional. The device type this token is used for. If not set, defaults to `ENROLLMENT_TOKEN_TYPE_CHROME_BROWSER`.
+}
+
+
+
+ list(parent, filter=None, pageSize=None, pageToken=None, x__xgafv=None)
+ Lists all browser device enrollment tokens.
+
+Args:
+ parent: string, Required. The parent resource where this enrollment token will be created. Format: customers/{customer} (required)
+ filter: string, Optional. Filter to apply to list results. The following fields can be used in the filter: * device_type (currently only 'CHROME_BROWSER' is supported) * token_state (supported values: 'ACTIVE', 'EXPIRED', 'REVOKED') * org_unit_path The full path of the org unit, such as /Montreal/Sales. * org_unit_id The obfuscated id of the org unit, not the full path.
+ pageSize: integer, Optional. Maximum number of results to return. Maximum and default are 100.
+ pageToken: string, Optional. Token to specify next page in the list.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Response that contains a list of tokens.
+ "enrollmentTokens": [ # List of Chrome device enrollment tokens.
+ { # Enrollment token for a chrome device.
+ "createTime": "A String", # Output only. The creation time of the token.
+ "creatorId": "A String", # Output only. The id of the user who created the token.
+ "customerId": "A String", # Output only. The obfuscated ID of the customer to whom this token is associated.
+ "name": "A String", # Identifier. The resource name of the enrollment token. Format: customers/{customer}/enrollmentTokens/{token_permanent_id}
+ "orgUnitId": "A String", # Optional. Obfuscated ID of the organization unit this token is associated with. If not set, the token is created for the root organization unit.
+ "revokeTime": "A String", # Output only. The revocation time of the token, null if the token is not revoked.
+ "revokerId": "A String", # Output only. The id of the user who revoked the token, null if the token is not revoked.
+ "state": "A String", # Output only. The state of the token.
+ "token": "A String", # Output only. The value of the token stored on the device and used to identify it to the management service handling the enrollment request.
+ "tokenPermanentId": "A String", # Output only. Unique identifier of the enrollment token used to access information on tokens.
+ "tokenType": "A String", # Optional. The device type this token is used for. If not set, defaults to `ENROLLMENT_TOKEN_TYPE_CHROME_BROWSER`.
+ },
+ ],
+ "nextPageToken": "A String", # Token to retrieve the next page of results, or empty if there are no more results in the list.
+}
+
+
+
+ list_next(previous_request, previous_response)
+ Retrieves the next page of results.
+
+Args:
+ previous_request: The request for the previous page. (required)
+ previous_response: The response from the request for the previous page. (required)
+
+Returns:
+ A request object that you can call 'execute()' on to request the next
+ page. Returns None if there are no more items in the collection.
+
+
+
+
+ revoke(name, body=None, x__xgafv=None)
+ Revokes a browser device enrollment token.
+
+Args:
+ name: string, Required. The name of the enrollment token to revoke. Format: customers/{customer}/enrollmentTokens/{token_permanent_id} (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request to revoke a token.
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Enrollment token for a chrome device.
+ "createTime": "A String", # Output only. The creation time of the token.
+ "creatorId": "A String", # Output only. The id of the user who created the token.
+ "customerId": "A String", # Output only. The obfuscated ID of the customer to whom this token is associated.
+ "name": "A String", # Identifier. The resource name of the enrollment token. Format: customers/{customer}/enrollmentTokens/{token_permanent_id}
+ "orgUnitId": "A String", # Optional. Obfuscated ID of the organization unit this token is associated with. If not set, the token is created for the root organization unit.
+ "revokeTime": "A String", # Output only. The revocation time of the token, null if the token is not revoked.
+ "revokerId": "A String", # Output only. The id of the user who revoked the token, null if the token is not revoked.
+ "state": "A String", # Output only. The state of the token.
+ "token": "A String", # Output only. The value of the token stored on the device and used to identify it to the management service handling the enrollment request.
+ "tokenPermanentId": "A String", # Output only. Unique identifier of the enrollment token used to access information on tokens.
+ "tokenType": "A String", # Optional. The device type this token is used for. If not set, defaults to `ENROLLMENT_TOKEN_TYPE_CHROME_BROWSER`.
+}
+
+
+
\ No newline at end of file
diff --git a/docs/dyn/chromemanagement_v1.customers.html b/docs/dyn/chromemanagement_v1.customers.html
index af370e4627..3668426a4d 100644
--- a/docs/dyn/chromemanagement_v1.customers.html
+++ b/docs/dyn/chromemanagement_v1.customers.html
@@ -84,11 +84,21 @@ Instance Methods
Returns the certificateProvisioningProcesses Resource.
+
+ chromeBrowsers()
+
+Returns the chromeBrowsers Resource.
+
Returns the connectorConfigs Resource.
+
+Returns the enrollmentTokens Resource.
+
diff --git a/docs/dyn/cloudlocationfinder_v1.projects.locations.cloudLocations.html b/docs/dyn/cloudlocationfinder_v1.projects.locations.cloudLocations.html
index dc245e347c..70428f3766 100644
--- a/docs/dyn/cloudlocationfinder_v1.projects.locations.cloudLocations.html
+++ b/docs/dyn/cloudlocationfinder_v1.projects.locations.cloudLocations.html
@@ -133,7 +133,7 @@ Method Details
Args:
parent: string, Required. The parent, which owns this collection of cloud locations. Format: projects/{project}/locations/{location} (required)
filter: string, Optional. A filter expression that filters resources listed in the response. The expression is in the form of field=value. For example, 'cloud_location_type=CLOUD_LOCATION_TYPE_REGION'. Multiple filter queries are space-separated. For example, 'cloud_location_type=CLOUD_LOCATION_TYPE_REGION territory_code="US"' By default, each expression is an AND expression. However, you can include AND and OR expressions explicitly.
- pageSize: integer, Optional. The maximum number of cloud locations to return per page. The service might return fewer cloud locations than this value. If unspecified, server will pick an appropriate default.
+ pageSize: integer, Optional. The maximum number of cloud locations to return per page. The service might return fewer cloud locations than this value. If unspecified, at most 500 cloud locations will be returned. The maximum value is 1000; values above 1000 will be coerced to 1000.
pageToken: string, Optional. A token identifying a page of results the server should return. Provide page token returned by a previous 'ListCloudLocations' call to retrieve the next page of results. When paginating, all other parameters provided to 'ListCloudLocations' must match the call that provided the page token.
x__xgafv: string, V1 error format.
Allowed values
@@ -182,7 +182,7 @@ Method Details
Args:
parent: string, Required. The parent, which owns this collection of cloud locations. Format: projects/{project}/locations/{location} (required)
- pageSize: integer, Optional. The maximum number of cloud locations to return. The service might return fewer cloud locations than this value. If unspecified, server will pick an appropriate default.
+ pageSize: integer, Optional. The maximum number of cloud locations to return. The service might return fewer cloud locations than this value. If unspecified, at most 500 cloud locations will be returned. The maximum value is 1000; values above 1000 will be coerced to 1000.
pageToken: string, Optional. A token identifying a page of results the server should return. Provide Page token returned by a previous 'ListCloudLocations' call to retrieve the next page of results. When paginating, all other parameters provided to 'ListCloudLocations' must match the call that provided the page token.
query: string, Optional. The query string in search query syntax. While filter is used to filter the search results by attributes, query is used to specify the search requirements.
sourceCloudLocation: string, Required. The source cloud location to search from. Example search can be searching nearby cloud locations from the source cloud location by latency.
diff --git a/docs/dyn/cloudlocationfinder_v1alpha.projects.locations.cloudLocations.html b/docs/dyn/cloudlocationfinder_v1alpha.projects.locations.cloudLocations.html
index 34ea380f75..2a103f87d2 100644
--- a/docs/dyn/cloudlocationfinder_v1alpha.projects.locations.cloudLocations.html
+++ b/docs/dyn/cloudlocationfinder_v1alpha.projects.locations.cloudLocations.html
@@ -133,7 +133,7 @@ Method Details
Args:
parent: string, Required. The parent, which owns this collection of cloud locations. Format: projects/{project}/locations/{location} (required)
filter: string, Optional. A filter expression that filters resources listed in the response. The expression is in the form of field=value. For example, 'cloud_location_type=CLOUD_LOCATION_TYPE_REGION'. Multiple filter queries are space-separated. For example, 'cloud_location_type=CLOUD_LOCATION_TYPE_REGION territory_code="US"' By default, each expression is an AND expression. However, you can include AND and OR expressions explicitly.
- pageSize: integer, Optional. The maximum number of cloud locations to return per page. The service might return fewer cloud locations than this value. If unspecified, server will pick an appropriate default.
+ pageSize: integer, Optional. The maximum number of cloud locations to return per page. The service might return fewer cloud locations than this value. If unspecified, at most 500 cloud locations will be returned. The maximum value is 1000; values above 1000 will be coerced to 1000.
pageToken: string, Optional. A token identifying a page of results the server should return. Provide page token returned by a previous 'ListCloudLocations' call to retrieve the next page of results. When paginating, all other parameters provided to 'ListCloudLocations' must match the call that provided the page token.
x__xgafv: string, V1 error format.
Allowed values
@@ -182,7 +182,7 @@ Method Details
Args:
parent: string, Required. The parent, which owns this collection of cloud locations. Format: projects/{project}/locations/{location} (required)
- pageSize: integer, Optional. The maximum number of cloud locations to return. The service might return fewer cloud locations than this value. If unspecified, server will pick an appropriate default.
+ pageSize: integer, Optional. The maximum number of cloud locations to return. The service might return fewer cloud locations than this value. If unspecified, at most 500 cloud locations will be returned. The maximum value is 1000; values above 1000 will be coerced to 1000.
pageToken: string, Optional. A token identifying a page of results the server should return. Provide Page token returned by a previous 'ListCloudLocations' call to retrieve the next page of results. When paginating, all other parameters provided to 'ListCloudLocations' must match the call that provided the page token.
query: string, Optional. The query string in search query syntax. While filter is used to filter the search results by attributes, query is used to specify the search requirements.
sourceCloudLocation: string, Required. The source cloud location to search from. Example search can be searching nearby cloud locations from the source cloud location by latency.
diff --git a/docs/dyn/compute_alpha.html b/docs/dyn/compute_alpha.html
index 48dee88a34..ba88139878 100644
--- a/docs/dyn/compute_alpha.html
+++ b/docs/dyn/compute_alpha.html
@@ -394,11 +394,6 @@ Instance Methods
Returns the nodeTypes Resource.
-
-Returns the orgVmExtensionPolicies Resource.
-
organizationGlobalVmExtensionPolicies()
diff --git a/docs/dyn/compute_alpha.organizationZoneVmExtensionPolicies.html b/docs/dyn/compute_alpha.organizationZoneVmExtensionPolicies.html
index c88037968b..76d35ada57 100644
--- a/docs/dyn/compute_alpha.organizationZoneVmExtensionPolicies.html
+++ b/docs/dyn/compute_alpha.organizationZoneVmExtensionPolicies.html
@@ -77,6 +77,9 @@ Instance Methods
Close httplib2 connections.
+
+ delete(organization, zone, vmExtensionPolicy, requestId=None, x__xgafv=None)
+Deletes a specified zone VM extension policy within an organization.
get(organization, zone, vmExtensionPolicy, x__xgafv=None)
Retrieves details of a specific zone VM extension policy within an
@@ -107,6 +110,317 @@ Method Details
Close httplib2 connections.
delete(organization, zone, vmExtensionPolicy, requestId=None, x__xgafv=None)
+ Deletes a specified zone VM extension policy within an organization.
+
+Args:
+ organization: string, Organization ID for this request. (required)
+ zone: string, Name of the zone for this request. (required)
+ vmExtensionPolicy: string, Name of the zone VM extension policy to delete. (required)
+ requestId: string, An optional request ID to identify requests. Specify a unique request ID so
+that if you must retry your request, the server will know to ignore the
+request if it has already been completed.
+
+For example, consider a situation where you make an initial request and
+the request times out. If you make the request again with the same
+request ID, the server can check if original operation with the same
+request ID was received, and if so, will ignore the second request. This
+prevents clients from accidentally creating duplicate commitments.
+
+The request ID must be
+a valid UUID with the exception that zero UUID is not supported
+(00000000-0000-0000-0000-000000000000).
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Represents an Operation resource.
+ #
+ # Google Compute Engine has three Operation resources:
+ #
+ # * [Global](/compute/docs/reference/rest/alpha/globalOperations)
+ # * [Regional](/compute/docs/reference/rest/alpha/regionOperations)
+ # * [Zonal](/compute/docs/reference/rest/alpha/zoneOperations)
+ #
+ # You can use an operation resource to manage asynchronous API requests.
+ # For more information, readHandling
+ # API responses.
+ #
+ # Operations can be global, regional or zonal.
+ #
+ # - For global operations, use the `globalOperations`
+ # resource.
+ # - For regional operations, use the
+ # `regionOperations` resource.
+ # - For zonal operations, use
+ # the `zoneOperations` resource.
+ #
+ #
+ #
+ # For more information, read
+ # Global, Regional, and Zonal Resources.
+ #
+ # Note that completed Operation resources have a limited
+ # retention period.
+ "clientOperationId": "A String", # [Output Only] The value of `requestId` if you provided it in the request.
+ # Not present otherwise.
+ "creationTimestamp": "A String", # [Deprecated] This field is deprecated.
+ "description": "A String", # [Output Only] A textual description of the operation, which is
+ # set when the operation is created.
+ "endTime": "A String", # [Output Only] The time that this operation was completed. This value is inRFC3339
+ # text format.
+ "error": { # [Output Only] If errors are generated during processing of the operation,
+ # this field will be populated.
+ "errors": [ # [Output Only] The array of errors encountered while processing this
+ # operation.
+ { # Represents a single error encountered during the processing of an
+ # operation.
+ "code": "A String", # [Output Only] The error type identifier for this error.
+ "errorDetails": [ # [Output Only] An optional list of messages that contain the error
+ # details. There is a set of defined message types to use for providing
+ # details.The syntax depends on the error code. For example,
+ # QuotaExceededInfo will have details when the error code is
+ # QUOTA_EXCEEDED.
+ { # Container for structured error details providing additional context
+ # specific to the encountered error code.
+ "errorInfo": { # Describes the cause of the error with structured details. # Error information containing structured domain, reason, and metadata.
+ #
+ # Example of an error when contacting the "pubsub.googleapis.com" API when it
+ # is not enabled:
+ #
+ # { "reason": "API_DISABLED"
+ # "domain": "googleapis.com"
+ # "metadata": {
+ # "resource": "projects/123",
+ # "service": "pubsub.googleapis.com"
+ # }
+ # }
+ #
+ # This response indicates that the pubsub.googleapis.com API is not enabled.
+ #
+ # Example of an error that is returned when attempting to create a Spanner
+ # instance in a region that is out of stock:
+ #
+ # { "reason": "STOCKOUT"
+ # "domain": "spanner.googleapis.com",
+ # "metadata": {
+ # "availableRegions": "us-central1,us-east2"
+ # }
+ # }
+ "domain": "A String", # The logical grouping to which the "reason" belongs. The error domain
+ # is typically the registered service name of the tool or product that
+ # generates the error. Example: "pubsub.googleapis.com". If the error is
+ # generated by some common infrastructure, the error domain must be a
+ # globally unique value that identifies the infrastructure. For Google API
+ # infrastructure, the error domain is "googleapis.com".
+ "metadatas": { # Additional structured details about this error.
+ #
+ # Keys must match a regular expression of `a-z+` but should
+ # ideally be lowerCamelCase. Also, they must be limited to 64 characters in
+ # length. When identifying the current value of an exceeded limit, the units
+ # should be contained in the key, not the value. For example, rather than
+ # `{"instanceLimit": "100/request"}`, should be returned as,
+ # `{"instanceLimitPerRequest": "100"}`, if the client exceeds the number of
+ # instances that can be created in a single (batch) request.
+ "a_key": "A String",
+ },
+ "reason": "A String", # The reason of the error. This is a constant value that identifies the
+ # proximate cause of the error. Error reasons are unique within a particular
+ # domain of errors. This should be at most 63 characters and match a
+ # regular expression of `A-Z+[A-Z0-9]`, which represents
+ # UPPER_SNAKE_CASE.
+ },
+ "help": { # Provides links to documentation or for performing an out of band action. # Links and information to help the user resolve the error.
+ #
+ # For example, if a quota check failed with an error indicating the calling
+ # project hasn't enabled the accessed service, this can contain a URL pointing
+ # directly to the right place in the developer console to flip the bit.
+ "links": [ # URL(s) pointing to additional information on handling the current error.
+ { # Describes a URL link.
+ "description": "A String", # Describes what the link offers.
+ "url": "A String", # The URL of the link.
+ },
+ ],
+ },
+ "localizedMessage": { # Provides a localized error message that is safe to return to the user # A localized human-readable error message intended for end users.
+ # which can be attached to an RPC error.
+ "locale": "A String", # The locale used following the specification defined at
+ # https://www.rfc-editor.org/rfc/bcp/bcp47.txt.
+ # Examples are: "en-US", "fr-CH", "es-MX"
+ "message": "A String", # The localized error message in the above locale.
+ },
+ "quotaInfo": { # Additional details for quota exceeded error for resource quota. # Details about quota limits and metrics when a quota is exceeded.
+ "dimensions": { # The map holding related quota dimensions.
+ "a_key": "A String",
+ },
+ "futureLimit": 3.14, # Future quota limit being rolled out. The limit's unit depends on the quota
+ # type or metric.
+ "limit": 3.14, # Current effective quota limit. The limit's unit depends on the quota type
+ # or metric.
+ "limitName": "A String", # The name of the quota limit.
+ "metricName": "A String", # The Compute Engine quota metric name.
+ "rolloutStatus": "A String", # Rollout status of the future quota limit.
+ },
+ },
+ ],
+ "location": "A String", # [Output Only] Indicates the field in the request that caused the error.
+ # This property is optional.
+ "message": "A String", # [Output Only] An optional, human-readable error message.
+ },
+ ],
+ },
+ "firewallPolicyRuleOperationMetadata": { # Metadata for FirewallPolicyRule operations. # Output only. [Output Only] Metadata containing the allocated priority from the
+ # networkFirewallPolicies.addRule and regionNetworkFirewallPolicies.addRule
+ # methods if not explicitly provided by the user.
+ "allocatedPriority": 42, # Output only. [Output Only] The priority allocated for the firewall policy rule if query
+ # parameters specified minPriority/maxPriority.
+ },
+ "getHealthOperationMetadata": { # Metadata for GetHealth operations. # Output only. Metadata for GetHealth operations.
+ "healthInfo": { # Health information. # Output only. The health information.
+ "availabilitySloStatus": "A String", # Output only. The availability SLO status.
+ "healthStatus": "A String", # Output only. The health status.
+ "repairCategory": "A String", # Output only. The repair category.
+ "unhealthyReason": "A String", # Output only. The reason for unhealthy status.
+ "updateTime": "A String", # Output only. The time when health info was updated.
+ },
+ },
+ "getVersionOperationMetadata": {
+ "inlineSbomInfo": {
+ "currentComponentVersions": { # A mapping of components to their currently-applied versions or other
+ # appropriate identifiers.
+ "a_key": "A String",
+ },
+ "targetComponentVersions": { # A mapping of components to their target versions or other appropriate
+ # identifiers.
+ "a_key": "A String",
+ },
+ },
+ },
+ "httpErrorMessage": "A String", # [Output Only] If the operation fails, this field contains the HTTP error
+ # message that was returned, such as `NOT FOUND`.
+ "httpErrorStatusCode": 42, # [Output Only] If the operation fails, this field contains the HTTP error
+ # status code that was returned. For example, a `404` means the
+ # resource was not found.
+ "id": "A String", # [Output Only] The unique identifier for the operation. This identifier is
+ # defined by the server.
+ "insertTime": "A String", # [Output Only] The time that this operation was requested.
+ # This value is inRFC3339
+ # text format.
+ "instancesBulkInsertOperationMetadata": {
+ "perLocationStatus": { # Status information per location (location name is key).
+ # Example key: zones/us-central1-a
+ "a_key": {
+ "createdVmCount": 42, # [Output Only] Count of VMs successfully created so far.
+ "deletedVmCount": 42, # [Output Only] Count of VMs that got deleted during rollback.
+ "failedToCreateVmCount": 42, # [Output Only] Count of VMs that started creating but encountered an
+ # error.
+ "status": "A String", # [Output Only] Creation status of BulkInsert operation - information
+ # if the flow is rolling forward or rolling back.
+ "targetVmCount": 42, # [Output Only] Count of VMs originally planned to be created.
+ },
+ },
+ },
+ "kind": "compute#operation", # Output only. [Output Only] Type of the resource. Always `compute#operation` for
+ # Operation resources.
+ "name": "A String", # [Output Only] Name of the operation.
+ "operationGroupId": "A String", # Output only. [Output Only] An ID that represents a group of operations, such as when a
+ # group of operations results from a `bulkInsert` API request.
+ "operationType": "A String", # [Output Only] The type of operation, such as `insert`,
+ # `update`, or `delete`, and so on.
+ "progress": 42, # [Output Only] An optional progress indicator that ranges from 0 to 100.
+ # There is no requirement that this be linear or support any granularity of
+ # operations. This should not be used to guess when the operation will be
+ # complete. This number should monotonically increase as the operation
+ # progresses.
+ "region": "A String", # [Output Only] The URL of the region where the operation resides. Only
+ # applicable when performing regional operations.
+ "selfLink": "A String", # [Output Only] Server-defined URL for the resource.
+ "selfLinkWithId": "A String", # Output only. [Output Only] Server-defined URL for this resource with the resource id.
+ "setCommonInstanceMetadataOperationMetadata": { # Output only. [Output Only] If the operation is for projects.setCommonInstanceMetadata,
+ # this field will contain information on all underlying zonal actions and
+ # their state.
+ "clientOperationId": "A String", # [Output Only] The client operation id.
+ "perLocationOperations": { # [Output Only] Status information per location (location name is key).
+ # Example key: zones/us-central1-a
+ "a_key": {
+ "error": { # The `Status` type defines a logical error model that is suitable for # [Output Only] If state is `ABANDONED` or `FAILED`, this field is
+ # populated.
+ # different programming environments, including REST APIs and RPC APIs. It is
+ # used by [gRPC](https://github.com/grpc). Each `Status` message contains
+ # three pieces of data: error code, error message, and error details.
+ #
+ # You can find out more about this error model and how to work with it in the
+ # [API Design Guide](https://cloud.google.com/apis/design/errors).
+ "code": 42, # The status code, which should be an enum value of google.rpc.Code.
+ "details": [ # A list of messages that carry the error details. There is a common set of
+ # message types for APIs to use.
+ {
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ ],
+ "message": "A String", # A developer-facing error message, which should be in English. Any
+ # user-facing error message should be localized and sent in the
+ # google.rpc.Status.details field, or localized by the client.
+ },
+ "state": "A String", # [Output Only] Status of the action, which can be one of the following:
+ # `PROPAGATING`, `PROPAGATED`, `ABANDONED`, `FAILED`, or `DONE`.
+ },
+ },
+ },
+ "startTime": "A String", # [Output Only] The time that this operation was started by the server.
+ # This value is inRFC3339
+ # text format.
+ "status": "A String", # [Output Only] The status of the operation, which can be one of the
+ # following:
+ # `PENDING`, `RUNNING`, or `DONE`.
+ "statusMessage": "A String", # [Output Only] An optional textual description of the current status of the
+ # operation.
+ "targetId": "A String", # [Output Only] The unique target ID, which identifies a specific incarnation
+ # of the target resource.
+ "targetLink": "A String", # [Output Only] The URL of the resource that the operation modifies. For
+ # operations related to creating a snapshot, this points to the disk
+ # that the snapshot was created from.
+ "user": "A String", # [Output Only] User who requested the operation, for example:
+ # `user@example.com` or
+ # `alice_smith_identifier (global/workforcePools/example-com-us-employees)`.
+ "warnings": [ # [Output Only] If warning messages are generated during processing of the
+ # operation, this field will be populated.
+ {
+ "code": "A String", # [Output Only] A warning code, if applicable. For example, Compute
+ # Engine returns NO_RESULTS_ON_PAGE if there
+ # are no results in the response.
+ "data": [ # [Output Only] Metadata about this warning in key:
+ # value format. For example:
+ #
+ # "data": [
+ # {
+ # "key": "scope",
+ # "value": "zones/us-east1-d"
+ # }]
+ {
+ "key": "A String", # [Output Only] A key that provides more detail on the warning being
+ # returned. For example, for warnings where there are no results in a list
+ # request for a particular zone, this key might be scope and
+ # the key value might be the zone name. Other examples might be a key
+ # indicating a deprecated resource and a suggested replacement, or a
+ # warning about invalid network settings (for example, if an instance
+ # attempts to perform IP forwarding but is not enabled for IP forwarding).
+ "value": "A String", # [Output Only] A warning data value corresponding to the key.
+ },
+ ],
+ "message": "A String", # [Output Only] A human-readable description of the warning code.
+ },
+ ],
+ "zone": "A String", # [Output Only] The URL of the zone where the operation resides. Only
+ # applicable when performing per-zone operations.
+}
+get(organization, zone, vmExtensionPolicy, x__xgafv=None)
Retrieves details of a specific zone VM extension policy within an diff --git a/docs/dyn/compute_preview.addresses.html b/docs/dyn/compute_preview.addresses.html index 6f4b9d6a58..9c7d4d78d7 100644 --- a/docs/dyn/compute_preview.addresses.html +++ b/docs/dyn/compute_preview.addresses.html @@ -293,6 +293,9 @@Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -336,6 +339,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and @@ -835,6 +841,9 @@Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -878,6 +887,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and @@ -980,6 +992,9 @@Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -1023,6 +1038,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and @@ -1508,6 +1526,9 @@Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -1551,6 +1572,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and diff --git a/docs/dyn/compute_preview.aliasNetworkEndpoints.html b/docs/dyn/compute_preview.aliasNetworkEndpoints.html new file mode 100644 index 0000000000..5bebd4a909 --- /dev/null +++ b/docs/dyn/compute_preview.aliasNetworkEndpoints.html @@ -0,0 +1,1022 @@ + + + +Compute Engine API . aliasNetworkEndpoints
+Instance Methods
++
+close()Close httplib2 connections.
++
+delete(project, zone, aliasNetworkEndpoint, requestId=None, x__xgafv=None)Deletes the specified AliasNetworkEndpoint.
++
+get(project, zone, aliasNetworkEndpoint, x__xgafv=None)Returns the specified AliasNetworkEndpoint resource.
++
+insert(project, zone, body=None, requestId=None, x__xgafv=None)Creates an AliasNetworkEndpoint in the specified project and zone
++
+list(project, zone, filter=None, maxResults=None, orderBy=None, pageToken=None, x__xgafv=None)Retrieves a list of AliasNetworkEndpoints available to the specified
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +delete(project, zone, aliasNetworkEndpoint, requestId=None, x__xgafv=None)+Deletes the specified AliasNetworkEndpoint. + +Args: + project: string, Required. Project ID for this request. (required) + zone: string, Required. Name of the zone scoping this request. (required) + aliasNetworkEndpoint: string, Required. Name of the AliasNetworkEndpoint resource to delete. (required) + requestId: string, An optional request ID to identify requests. Specify a unique request ID so +that if you must retry your request, the server will know to ignore the +request if it has already been completed. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an Operation resource. + # + # Google Compute Engine has three Operation resources: + # + # * [Global](/compute/docs/reference/rest/v1/globalOperations) + # * [Regional](/compute/docs/reference/rest/v1/regionOperations) + # * [Zonal](/compute/docs/reference/rest/v1/zoneOperations) + # + # You can use an operation resource to manage asynchronous API requests. + # For more information, readHandling + # API responses. + # + # Operations can be global, regional or zonal. + # + # - For global operations, use the `globalOperations` + # resource. + # - For regional operations, use the + # `regionOperations` resource. + # - For zonal operations, use + # the `zoneOperations` resource. + # + # + # + # For more information, read + # Global, Regional, and Zonal Resources. + # + # Note that completed Operation resources have a limited + # retention period. + "clientOperationId": "A String", # [Output Only] The value of `requestId` if you provided it in the request. + # Not present otherwise. + "creationTimestamp": "A String", # [Deprecated] This field is deprecated. + "description": "A String", # [Output Only] A textual description of the operation, which is + # set when the operation is created. + "endTime": "A String", # [Output Only] The time that this operation was completed. This value is inRFC3339 + # text format. + "error": { # [Output Only] If errors are generated during processing of the operation, + # this field will be populated. + "errors": [ # [Output Only] The array of errors encountered while processing this + # operation. + { # Represents a single error encountered during the processing of an + # operation. + "code": "A String", # [Output Only] The error type identifier for this error. + "errorDetails": [ # [Output Only] An optional list of messages that contain the error + # details. There is a set of defined message types to use for providing + # details.The syntax depends on the error code. For example, + # QuotaExceededInfo will have details when the error code is + # QUOTA_EXCEEDED. + { # Container for structured error details providing additional context + # specific to the encountered error code. + "errorInfo": { # Describes the cause of the error with structured details. # Error information containing structured domain, reason, and metadata. + # + # Example of an error when contacting the "pubsub.googleapis.com" API when it + # is not enabled: + # + # { "reason": "API_DISABLED" + # "domain": "googleapis.com" + # "metadata": { + # "resource": "projects/123", + # "service": "pubsub.googleapis.com" + # } + # } + # + # This response indicates that the pubsub.googleapis.com API is not enabled. + # + # Example of an error that is returned when attempting to create a Spanner + # instance in a region that is out of stock: + # + # { "reason": "STOCKOUT" + # "domain": "spanner.googleapis.com", + # "metadata": { + # "availableRegions": "us-central1,us-east2" + # } + # } + "domain": "A String", # The logical grouping to which the "reason" belongs. The error domain + # is typically the registered service name of the tool or product that + # generates the error. Example: "pubsub.googleapis.com". If the error is + # generated by some common infrastructure, the error domain must be a + # globally unique value that identifies the infrastructure. For Google API + # infrastructure, the error domain is "googleapis.com". + "metadatas": { # Additional structured details about this error. + # + # Keys must match a regular expression of `a-z+` but should + # ideally be lowerCamelCase. Also, they must be limited to 64 characters in + # length. When identifying the current value of an exceeded limit, the units + # should be contained in the key, not the value. For example, rather than + # `{"instanceLimit": "100/request"}`, should be returned as, + # `{"instanceLimitPerRequest": "100"}`, if the client exceeds the number of + # instances that can be created in a single (batch) request. + "a_key": "A String", + }, + "reason": "A String", # The reason of the error. This is a constant value that identifies the + # proximate cause of the error. Error reasons are unique within a particular + # domain of errors. This should be at most 63 characters and match a + # regular expression of `A-Z+[A-Z0-9]`, which represents + # UPPER_SNAKE_CASE. + }, + "help": { # Provides links to documentation or for performing an out of band action. # Links and information to help the user resolve the error. + # + # For example, if a quota check failed with an error indicating the calling + # project hasn't enabled the accessed service, this can contain a URL pointing + # directly to the right place in the developer console to flip the bit. + "links": [ # URL(s) pointing to additional information on handling the current error. + { # Describes a URL link. + "description": "A String", # Describes what the link offers. + "url": "A String", # The URL of the link. + }, + ], + }, + "localizedMessage": { # Provides a localized error message that is safe to return to the user # A localized human-readable error message intended for end users. + # which can be attached to an RPC error. + "locale": "A String", # The locale used following the specification defined at + # https://www.rfc-editor.org/rfc/bcp/bcp47.txt. + # Examples are: "en-US", "fr-CH", "es-MX" + "message": "A String", # The localized error message in the above locale. + }, + "quotaInfo": { # Additional details for quota exceeded error for resource quota. # Details about quota limits and metrics when a quota is exceeded. + "dimensions": { # The map holding related quota dimensions. + "a_key": "A String", + }, + "futureLimit": 3.14, # Future quota limit being rolled out. The limit's unit depends on the quota + # type or metric. + "limit": 3.14, # Current effective quota limit. The limit's unit depends on the quota type + # or metric. + "limitName": "A String", # The name of the quota limit. + "metricName": "A String", # The Compute Engine quota metric name. + "rolloutStatus": "A String", # Rollout status of the future quota limit. + }, + }, + ], + "location": "A String", # [Output Only] Indicates the field in the request that caused the error. + # This property is optional. + "message": "A String", # [Output Only] An optional, human-readable error message. + }, + ], + }, + "getHealthOperationMetadata": { # Metadata for GetHealth operations. # Output only. Metadata for GetHealth operations. + "healthInfo": { # Health information. # Output only. The health information. + "availabilitySloStatus": "A String", # Output only. The availability SLO status. + "healthStatus": "A String", # Output only. The health status. + "repairCategory": "A String", # Output only. The repair category. + "unhealthyReason": "A String", # Output only. The reason for unhealthy status. + "updateTime": "A String", # Output only. The time when health info was updated. + }, + }, + "getVersionOperationMetadata": { + "inlineSbomInfo": { + "currentComponentVersions": { # A mapping of components to their currently-applied versions or other + # appropriate identifiers. + "a_key": "A String", + }, + "targetComponentVersions": { # A mapping of components to their target versions or other appropriate + # identifiers. + "a_key": "A String", + }, + }, + }, + "httpErrorMessage": "A String", # [Output Only] If the operation fails, this field contains the HTTP error + # message that was returned, such as `NOT FOUND`. + "httpErrorStatusCode": 42, # [Output Only] If the operation fails, this field contains the HTTP error + # status code that was returned. For example, a `404` means the + # resource was not found. + "id": "A String", # [Output Only] The unique identifier for the operation. This identifier is + # defined by the server. + "insertTime": "A String", # [Output Only] The time that this operation was requested. + # This value is inRFC3339 + # text format. + "instancesBulkInsertOperationMetadata": { + "perLocationStatus": { # Status information per location (location name is key). + # Example key: zones/us-central1-a + "a_key": { + "createdVmCount": 42, # [Output Only] Count of VMs successfully created so far. + "deletedVmCount": 42, # [Output Only] Count of VMs that got deleted during rollback. + "failedToCreateVmCount": 42, # [Output Only] Count of VMs that started creating but encountered an + # error. + "status": "A String", # [Output Only] Creation status of BulkInsert operation - information + # if the flow is rolling forward or rolling back. + "targetVmCount": 42, # [Output Only] Count of VMs originally planned to be created. + }, + }, + }, + "instancesTroubleshootOperationMetadata": { # [Output Only] Operation metadata for instances.troubleshoot. # Output only. [Output Only] Operation metadata for instances.troubleshoot. + "troubleshootOutput": "A String", # Output only. [Output Only] Serialized output of the troubleshooting diagnostic run. + }, + "kind": "compute#operation", # Output only. [Output Only] Type of the resource. Always `compute#operation` for + # Operation resources. + "name": "A String", # [Output Only] Name of the operation. + "operationGroupId": "A String", # Output only. [Output Only] An ID that represents a group of operations, such as when a + # group of operations results from a `bulkInsert` API request. + "operationType": "A String", # [Output Only] The type of operation, such as `insert`, + # `update`, or `delete`, and so on. + "progress": 42, # [Output Only] An optional progress indicator that ranges from 0 to 100. + # There is no requirement that this be linear or support any granularity of + # operations. This should not be used to guess when the operation will be + # complete. This number should monotonically increase as the operation + # progresses. + "region": "A String", # [Output Only] The URL of the region where the operation resides. Only + # applicable when performing regional operations. + "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "setCommonInstanceMetadataOperationMetadata": { # Output only. [Output Only] If the operation is for projects.setCommonInstanceMetadata, + # this field will contain information on all underlying zonal actions and + # their state. + "clientOperationId": "A String", # [Output Only] The client operation id. + "perLocationOperations": { # [Output Only] Status information per location (location name is key). + # Example key: zones/us-central1-a + "a_key": { + "error": { # The `Status` type defines a logical error model that is suitable for # [Output Only] If state is `ABANDONED` or `FAILED`, this field is + # populated. + # different programming environments, including REST APIs and RPC APIs. It is + # used by [gRPC](https://github.com/grpc). Each `Status` message contains + # three pieces of data: error code, error message, and error details. + # + # You can find out more about this error model and how to work with it in the + # [API Design Guide](https://cloud.google.com/apis/design/errors). + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of + # message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any + # user-facing error message should be localized and sent in the + # google.rpc.Status.details field, or localized by the client. + }, + "state": "A String", # [Output Only] Status of the action, which can be one of the following: + # `PROPAGATING`, `PROPAGATED`, `ABANDONED`, `FAILED`, or `DONE`. + }, + }, + }, + "startTime": "A String", # [Output Only] The time that this operation was started by the server. + # This value is inRFC3339 + # text format. + "status": "A String", # [Output Only] The status of the operation, which can be one of the + # following: + # `PENDING`, `RUNNING`, or `DONE`. + "statusMessage": "A String", # [Output Only] An optional textual description of the current status of the + # operation. + "targetId": "A String", # [Output Only] The unique target ID, which identifies a specific incarnation + # of the target resource. + "targetLink": "A String", # [Output Only] The URL of the resource that the operation modifies. For + # operations related to creating a snapshot, this points to the disk + # that the snapshot was created from. + "user": "A String", # [Output Only] User who requested the operation, for example: + # `user@example.com` or + # `alice_smith_identifier (global/workforcePools/example-com-us-employees)`. + "warnings": [ # [Output Only] If warning messages are generated during processing of the + # operation, this field will be populated. + { + "code": "A String", # [Output Only] A warning code, if applicable. For example, Compute + # Engine returns NO_RESULTS_ON_PAGE if there + # are no results in the response. + "data": [ # [Output Only] Metadata about this warning in key: + # value format. For example: + # + # "data": [ + # { + # "key": "scope", + # "value": "zones/us-east1-d" + # }] + { + "key": "A String", # [Output Only] A key that provides more detail on the warning being + # returned. For example, for warnings where there are no results in a list + # request for a particular zone, this key might be scope and + # the key value might be the zone name. Other examples might be a key + # indicating a deprecated resource and a suggested replacement, or a + # warning about invalid network settings (for example, if an instance + # attempts to perform IP forwarding but is not enabled for IP forwarding). + "value": "A String", # [Output Only] A warning data value corresponding to the key. + }, + ], + "message": "A String", # [Output Only] A human-readable description of the warning code. + }, + ], + "zone": "A String", # [Output Only] The URL of the zone where the operation resides. Only + # applicable when performing per-zone operations. +}+++ +get(project, zone, aliasNetworkEndpoint, x__xgafv=None)+Returns the specified AliasNetworkEndpoint resource. + +Args: + project: string, Required. Project ID for this request. (required) + zone: string, Required. Name of the zone scoping this request. (required) + aliasNetworkEndpoint: string, Required. Name of the AliasNetworkEndpoint resource to return. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an Alias Network Endpoint resource. + # To be used for GKE pod-native endpoints in Andromeda. + "aliases": { # Required. The map of network aliases allocated for the AliasNetworkEndpoint. + # This map is keyed with user-provided alias names. + "a_key": { # Configuration for an IP alias. + "effectiveIpAddress": "A String", # Output only. [Output Only] The actual IP address allocated by the system. + "effectiveSubnetworkRangeName": "A String", # Output only. [Output Only] The actual secondary range used for allocation. + "ipAddress": "A String", # Input only. The IP address of the network endpoint. If specified, the server performs + # a validation check for uniqueness. If the field is omitted, the system + # automatically allocates an IP and populates the effective_ip_address + # field. + "ipVersion": "A String", # Required. The IP version of the IP. It is IPv4 or IPv6. + "subnetworkRangeNames": [ # Input only. Defines candidate secondary ranges for allocating the alias IPs. If it is + # empty, the IP would be allocated from the primary range of the + # subnetwork. + # This field is INPUT_ONLY and used only during resource creation, + # hence a repeated field is acceptable as per AIP-2825. It is not mutable. + "A String", + ], + }, + }, + "annotations": { # Optional. The annotations map for the AliasNetworkEndpoint. + # Keys and values must follow Kubernetes limits. + # Keys must be a maximum of 253 characters. Keys should be dot-namespaced, + # following the Kubernetes annotation key format: + # `([DNS subdomain]/)?[RFC 1123 name]`. + # Values must be a maximum of 256 characters. + "a_key": "A String", + }, + "creationTimestamp": "A String", # Output only. [Output Only] Creation timestamp in RFC3339 text format. + "description": "A String", # Optional. An optional description of this resource. Provide this field when you + # create the resource. + "host": { # Information about the host instance. # Required. The host that the alias network endpoint is bound to. + "instance": "A String", # Required. The URL to the instance that this alias network endpoint is bound to. + }, + "id": "A String", # Output only. [Output Only] The unique identifier for the resource. This identifier is + # defined by the server. + "labels": { # Optional. The labels map for the AliasNetworkEndpoint. + "a_key": "A String", + }, + "name": "A String", # Identifier. Name of the resource. Provided by the client when the resource is created. + # The name must be 1-63 characters long, and comply with RFC1035. + "securityTagInheritance": "A String", # Optional. The security tag inheritance setting for the AliasNetworkEndpoint. + "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "status": { # Represents the current status of the AliasNetworkEndpoint. # Output only. [Output Only] Current status of the alias network endpoint. + "network": "A String", # Output only. [Output Only] The URL of the network that this alias network endpoint + # belongs to. + "state": "A String", # Output only. [Output Only] The state of the alias network endpoint. + }, + "subnetwork": "A String", # Required. The URL of the subnetwork that this alias network endpoint belongs to. + "updateTime": "A String", # Output only. [Output Only] Update timestamp in RFC3339 text format. +}+++ +insert(project, zone, body=None, requestId=None, x__xgafv=None)+Creates an AliasNetworkEndpoint in the specified project and zone +using the data included in the request. + +Args: + project: string, Required. Project ID for this request. (required) + zone: string, Required. Name of the zone scoping this request. (required) + body: object, The request body. + The object takes the form of: + +{ # Represents an Alias Network Endpoint resource. + # To be used for GKE pod-native endpoints in Andromeda. + "aliases": { # Required. The map of network aliases allocated for the AliasNetworkEndpoint. + # This map is keyed with user-provided alias names. + "a_key": { # Configuration for an IP alias. + "effectiveIpAddress": "A String", # Output only. [Output Only] The actual IP address allocated by the system. + "effectiveSubnetworkRangeName": "A String", # Output only. [Output Only] The actual secondary range used for allocation. + "ipAddress": "A String", # Input only. The IP address of the network endpoint. If specified, the server performs + # a validation check for uniqueness. If the field is omitted, the system + # automatically allocates an IP and populates the effective_ip_address + # field. + "ipVersion": "A String", # Required. The IP version of the IP. It is IPv4 or IPv6. + "subnetworkRangeNames": [ # Input only. Defines candidate secondary ranges for allocating the alias IPs. If it is + # empty, the IP would be allocated from the primary range of the + # subnetwork. + # This field is INPUT_ONLY and used only during resource creation, + # hence a repeated field is acceptable as per AIP-2825. It is not mutable. + "A String", + ], + }, + }, + "annotations": { # Optional. The annotations map for the AliasNetworkEndpoint. + # Keys and values must follow Kubernetes limits. + # Keys must be a maximum of 253 characters. Keys should be dot-namespaced, + # following the Kubernetes annotation key format: + # `([DNS subdomain]/)?[RFC 1123 name]`. + # Values must be a maximum of 256 characters. + "a_key": "A String", + }, + "creationTimestamp": "A String", # Output only. [Output Only] Creation timestamp in RFC3339 text format. + "description": "A String", # Optional. An optional description of this resource. Provide this field when you + # create the resource. + "host": { # Information about the host instance. # Required. The host that the alias network endpoint is bound to. + "instance": "A String", # Required. The URL to the instance that this alias network endpoint is bound to. + }, + "id": "A String", # Output only. [Output Only] The unique identifier for the resource. This identifier is + # defined by the server. + "labels": { # Optional. The labels map for the AliasNetworkEndpoint. + "a_key": "A String", + }, + "name": "A String", # Identifier. Name of the resource. Provided by the client when the resource is created. + # The name must be 1-63 characters long, and comply with RFC1035. + "securityTagInheritance": "A String", # Optional. The security tag inheritance setting for the AliasNetworkEndpoint. + "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "status": { # Represents the current status of the AliasNetworkEndpoint. # Output only. [Output Only] Current status of the alias network endpoint. + "network": "A String", # Output only. [Output Only] The URL of the network that this alias network endpoint + # belongs to. + "state": "A String", # Output only. [Output Only] The state of the alias network endpoint. + }, + "subnetwork": "A String", # Required. The URL of the subnetwork that this alias network endpoint belongs to. + "updateTime": "A String", # Output only. [Output Only] Update timestamp in RFC3339 text format. +} + + requestId: string, An optional request ID to identify requests. Specify a unique request ID so +that if you must retry your request, the server will know to ignore the +request if it has already been completed. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Represents an Operation resource. + # + # Google Compute Engine has three Operation resources: + # + # * [Global](/compute/docs/reference/rest/v1/globalOperations) + # * [Regional](/compute/docs/reference/rest/v1/regionOperations) + # * [Zonal](/compute/docs/reference/rest/v1/zoneOperations) + # + # You can use an operation resource to manage asynchronous API requests. + # For more information, readHandling + # API responses. + # + # Operations can be global, regional or zonal. + # + # - For global operations, use the `globalOperations` + # resource. + # - For regional operations, use the + # `regionOperations` resource. + # - For zonal operations, use + # the `zoneOperations` resource. + # + # + # + # For more information, read + # Global, Regional, and Zonal Resources. + # + # Note that completed Operation resources have a limited + # retention period. + "clientOperationId": "A String", # [Output Only] The value of `requestId` if you provided it in the request. + # Not present otherwise. + "creationTimestamp": "A String", # [Deprecated] This field is deprecated. + "description": "A String", # [Output Only] A textual description of the operation, which is + # set when the operation is created. + "endTime": "A String", # [Output Only] The time that this operation was completed. This value is inRFC3339 + # text format. + "error": { # [Output Only] If errors are generated during processing of the operation, + # this field will be populated. + "errors": [ # [Output Only] The array of errors encountered while processing this + # operation. + { # Represents a single error encountered during the processing of an + # operation. + "code": "A String", # [Output Only] The error type identifier for this error. + "errorDetails": [ # [Output Only] An optional list of messages that contain the error + # details. There is a set of defined message types to use for providing + # details.The syntax depends on the error code. For example, + # QuotaExceededInfo will have details when the error code is + # QUOTA_EXCEEDED. + { # Container for structured error details providing additional context + # specific to the encountered error code. + "errorInfo": { # Describes the cause of the error with structured details. # Error information containing structured domain, reason, and metadata. + # + # Example of an error when contacting the "pubsub.googleapis.com" API when it + # is not enabled: + # + # { "reason": "API_DISABLED" + # "domain": "googleapis.com" + # "metadata": { + # "resource": "projects/123", + # "service": "pubsub.googleapis.com" + # } + # } + # + # This response indicates that the pubsub.googleapis.com API is not enabled. + # + # Example of an error that is returned when attempting to create a Spanner + # instance in a region that is out of stock: + # + # { "reason": "STOCKOUT" + # "domain": "spanner.googleapis.com", + # "metadata": { + # "availableRegions": "us-central1,us-east2" + # } + # } + "domain": "A String", # The logical grouping to which the "reason" belongs. The error domain + # is typically the registered service name of the tool or product that + # generates the error. Example: "pubsub.googleapis.com". If the error is + # generated by some common infrastructure, the error domain must be a + # globally unique value that identifies the infrastructure. For Google API + # infrastructure, the error domain is "googleapis.com". + "metadatas": { # Additional structured details about this error. + # + # Keys must match a regular expression of `a-z+` but should + # ideally be lowerCamelCase. Also, they must be limited to 64 characters in + # length. When identifying the current value of an exceeded limit, the units + # should be contained in the key, not the value. For example, rather than + # `{"instanceLimit": "100/request"}`, should be returned as, + # `{"instanceLimitPerRequest": "100"}`, if the client exceeds the number of + # instances that can be created in a single (batch) request. + "a_key": "A String", + }, + "reason": "A String", # The reason of the error. This is a constant value that identifies the + # proximate cause of the error. Error reasons are unique within a particular + # domain of errors. This should be at most 63 characters and match a + # regular expression of `A-Z+[A-Z0-9]`, which represents + # UPPER_SNAKE_CASE. + }, + "help": { # Provides links to documentation or for performing an out of band action. # Links and information to help the user resolve the error. + # + # For example, if a quota check failed with an error indicating the calling + # project hasn't enabled the accessed service, this can contain a URL pointing + # directly to the right place in the developer console to flip the bit. + "links": [ # URL(s) pointing to additional information on handling the current error. + { # Describes a URL link. + "description": "A String", # Describes what the link offers. + "url": "A String", # The URL of the link. + }, + ], + }, + "localizedMessage": { # Provides a localized error message that is safe to return to the user # A localized human-readable error message intended for end users. + # which can be attached to an RPC error. + "locale": "A String", # The locale used following the specification defined at + # https://www.rfc-editor.org/rfc/bcp/bcp47.txt. + # Examples are: "en-US", "fr-CH", "es-MX" + "message": "A String", # The localized error message in the above locale. + }, + "quotaInfo": { # Additional details for quota exceeded error for resource quota. # Details about quota limits and metrics when a quota is exceeded. + "dimensions": { # The map holding related quota dimensions. + "a_key": "A String", + }, + "futureLimit": 3.14, # Future quota limit being rolled out. The limit's unit depends on the quota + # type or metric. + "limit": 3.14, # Current effective quota limit. The limit's unit depends on the quota type + # or metric. + "limitName": "A String", # The name of the quota limit. + "metricName": "A String", # The Compute Engine quota metric name. + "rolloutStatus": "A String", # Rollout status of the future quota limit. + }, + }, + ], + "location": "A String", # [Output Only] Indicates the field in the request that caused the error. + # This property is optional. + "message": "A String", # [Output Only] An optional, human-readable error message. + }, + ], + }, + "getHealthOperationMetadata": { # Metadata for GetHealth operations. # Output only. Metadata for GetHealth operations. + "healthInfo": { # Health information. # Output only. The health information. + "availabilitySloStatus": "A String", # Output only. The availability SLO status. + "healthStatus": "A String", # Output only. The health status. + "repairCategory": "A String", # Output only. The repair category. + "unhealthyReason": "A String", # Output only. The reason for unhealthy status. + "updateTime": "A String", # Output only. The time when health info was updated. + }, + }, + "getVersionOperationMetadata": { + "inlineSbomInfo": { + "currentComponentVersions": { # A mapping of components to their currently-applied versions or other + # appropriate identifiers. + "a_key": "A String", + }, + "targetComponentVersions": { # A mapping of components to their target versions or other appropriate + # identifiers. + "a_key": "A String", + }, + }, + }, + "httpErrorMessage": "A String", # [Output Only] If the operation fails, this field contains the HTTP error + # message that was returned, such as `NOT FOUND`. + "httpErrorStatusCode": 42, # [Output Only] If the operation fails, this field contains the HTTP error + # status code that was returned. For example, a `404` means the + # resource was not found. + "id": "A String", # [Output Only] The unique identifier for the operation. This identifier is + # defined by the server. + "insertTime": "A String", # [Output Only] The time that this operation was requested. + # This value is inRFC3339 + # text format. + "instancesBulkInsertOperationMetadata": { + "perLocationStatus": { # Status information per location (location name is key). + # Example key: zones/us-central1-a + "a_key": { + "createdVmCount": 42, # [Output Only] Count of VMs successfully created so far. + "deletedVmCount": 42, # [Output Only] Count of VMs that got deleted during rollback. + "failedToCreateVmCount": 42, # [Output Only] Count of VMs that started creating but encountered an + # error. + "status": "A String", # [Output Only] Creation status of BulkInsert operation - information + # if the flow is rolling forward or rolling back. + "targetVmCount": 42, # [Output Only] Count of VMs originally planned to be created. + }, + }, + }, + "instancesTroubleshootOperationMetadata": { # [Output Only] Operation metadata for instances.troubleshoot. # Output only. [Output Only] Operation metadata for instances.troubleshoot. + "troubleshootOutput": "A String", # Output only. [Output Only] Serialized output of the troubleshooting diagnostic run. + }, + "kind": "compute#operation", # Output only. [Output Only] Type of the resource. Always `compute#operation` for + # Operation resources. + "name": "A String", # [Output Only] Name of the operation. + "operationGroupId": "A String", # Output only. [Output Only] An ID that represents a group of operations, such as when a + # group of operations results from a `bulkInsert` API request. + "operationType": "A String", # [Output Only] The type of operation, such as `insert`, + # `update`, or `delete`, and so on. + "progress": 42, # [Output Only] An optional progress indicator that ranges from 0 to 100. + # There is no requirement that this be linear or support any granularity of + # operations. This should not be used to guess when the operation will be + # complete. This number should monotonically increase as the operation + # progresses. + "region": "A String", # [Output Only] The URL of the region where the operation resides. Only + # applicable when performing regional operations. + "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "setCommonInstanceMetadataOperationMetadata": { # Output only. [Output Only] If the operation is for projects.setCommonInstanceMetadata, + # this field will contain information on all underlying zonal actions and + # their state. + "clientOperationId": "A String", # [Output Only] The client operation id. + "perLocationOperations": { # [Output Only] Status information per location (location name is key). + # Example key: zones/us-central1-a + "a_key": { + "error": { # The `Status` type defines a logical error model that is suitable for # [Output Only] If state is `ABANDONED` or `FAILED`, this field is + # populated. + # different programming environments, including REST APIs and RPC APIs. It is + # used by [gRPC](https://github.com/grpc). Each `Status` message contains + # three pieces of data: error code, error message, and error details. + # + # You can find out more about this error model and how to work with it in the + # [API Design Guide](https://cloud.google.com/apis/design/errors). + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of + # message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any + # user-facing error message should be localized and sent in the + # google.rpc.Status.details field, or localized by the client. + }, + "state": "A String", # [Output Only] Status of the action, which can be one of the following: + # `PROPAGATING`, `PROPAGATED`, `ABANDONED`, `FAILED`, or `DONE`. + }, + }, + }, + "startTime": "A String", # [Output Only] The time that this operation was started by the server. + # This value is inRFC3339 + # text format. + "status": "A String", # [Output Only] The status of the operation, which can be one of the + # following: + # `PENDING`, `RUNNING`, or `DONE`. + "statusMessage": "A String", # [Output Only] An optional textual description of the current status of the + # operation. + "targetId": "A String", # [Output Only] The unique target ID, which identifies a specific incarnation + # of the target resource. + "targetLink": "A String", # [Output Only] The URL of the resource that the operation modifies. For + # operations related to creating a snapshot, this points to the disk + # that the snapshot was created from. + "user": "A String", # [Output Only] User who requested the operation, for example: + # `user@example.com` or + # `alice_smith_identifier (global/workforcePools/example-com-us-employees)`. + "warnings": [ # [Output Only] If warning messages are generated during processing of the + # operation, this field will be populated. + { + "code": "A String", # [Output Only] A warning code, if applicable. For example, Compute + # Engine returns NO_RESULTS_ON_PAGE if there + # are no results in the response. + "data": [ # [Output Only] Metadata about this warning in key: + # value format. For example: + # + # "data": [ + # { + # "key": "scope", + # "value": "zones/us-east1-d" + # }] + { + "key": "A String", # [Output Only] A key that provides more detail on the warning being + # returned. For example, for warnings where there are no results in a list + # request for a particular zone, this key might be scope and + # the key value might be the zone name. Other examples might be a key + # indicating a deprecated resource and a suggested replacement, or a + # warning about invalid network settings (for example, if an instance + # attempts to perform IP forwarding but is not enabled for IP forwarding). + "value": "A String", # [Output Only] A warning data value corresponding to the key. + }, + ], + "message": "A String", # [Output Only] A human-readable description of the warning code. + }, + ], + "zone": "A String", # [Output Only] The URL of the zone where the operation resides. Only + # applicable when performing per-zone operations. +}+++ +list(project, zone, filter=None, maxResults=None, orderBy=None, pageToken=None, x__xgafv=None)+Retrieves a list of AliasNetworkEndpoints available to the specified +project. + +Args: + project: string, Required. Project ID for this request. (required) + zone: string, Required. Name of the zone of this request. (required) + filter: string, A filter expression that filters resources listed in the response. Most +Compute resources support two types of filter expressions: +expressions that support regular expressions and expressions that follow +API improvement proposal AIP-160. +These two types of filter expressions cannot be mixed in one request. + +If you want to use AIP-160, your expression must specify the field name, an +operator, and the value that you want to use for filtering. The value +must be a string, a number, or a boolean. The operator +must be either `=`, `!=`, `>`, `<`, `<=`, `>=` or `:`. + +For example, if you are filtering Compute Engine instances, you can +exclude instances named `example-instance` by specifying +`name != example-instance`. + +The `:*` comparison can be used to test whether a key has been defined. +For example, to find all objects with `owner` label use: +``` +labels.owner:* +``` + +You can also filter nested fields. For example, you could specify +`scheduling.automaticRestart = false` to include instances only +if they are not scheduled for automatic restarts. You can use filtering +on nested fields to filter based onresource labels. + +To filter on multiple expressions, provide each separate expression within +parentheses. For example: +``` +(scheduling.automaticRestart = true) +(cpuPlatform = "Intel Skylake") +``` +By default, each expression is an `AND` expression. However, you +can include `AND` and `OR` expressions explicitly. +For example: +``` +(cpuPlatform = "Intel Skylake") OR +(cpuPlatform = "Intel Broadwell") AND +(scheduling.automaticRestart = true) +``` + +If you want to use a regular expression, use the `eq` (equal) or `ne` +(not equal) operator against a single un-parenthesized expression with or +without quotes or against multiple parenthesized expressions. Examples: + +`fieldname eq unquoted literal` +`fieldname eq 'single quoted literal'` +`fieldname eq "double quoted literal"` +`(fieldname1 eq literal) (fieldname2 ne "literal")` + +The literal value is interpreted as a regular expression using GoogleRE2 library syntax. +The literal value must match the entire field. + +For example, to filter for instances that do not end with name "instance", +you would use `name ne .*instance`. + +You cannot combine constraints on multiple fields using regular +expressions. + maxResults: integer, The maximum number of results per page that should be returned. +If the number of available results is larger than `maxResults`, +Compute Engine returns a `nextPageToken` that can be used to get +the next page of results in subsequent list requests. Acceptable values are +`0` to `500`, inclusive. (Default: `500`) + orderBy: string, Sorts list results by a certain order. By default, results +are returned in alphanumerical order based on the resource name. + +You can also sort results in descending order based on the creation +timestamp using `orderBy="creationTimestamp desc"`. This sorts +results based on the `creationTimestamp` field in +reverse chronological order (newest result first). Use this to sort +resources like operations so that the newest operation is returned first. + +Currently, only sorting by `name` or +`creationTimestamp desc` is supported. + pageToken: string, Specifies a page token to use. Set `pageToken` to the +`nextPageToken` returned by a previous list request to get +the next page of results. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Response message for AliasNetworkEndpoints.List. + "etag": "A String", # ETag of the resource list. + "id": "A String", # [Output Only] Unique identifier for the resource; defined by the server. + "items": [ # A list of AliasNetworkEndpoint resources. + { # Represents an Alias Network Endpoint resource. + # To be used for GKE pod-native endpoints in Andromeda. + "aliases": { # Required. The map of network aliases allocated for the AliasNetworkEndpoint. + # This map is keyed with user-provided alias names. + "a_key": { # Configuration for an IP alias. + "effectiveIpAddress": "A String", # Output only. [Output Only] The actual IP address allocated by the system. + "effectiveSubnetworkRangeName": "A String", # Output only. [Output Only] The actual secondary range used for allocation. + "ipAddress": "A String", # Input only. The IP address of the network endpoint. If specified, the server performs + # a validation check for uniqueness. If the field is omitted, the system + # automatically allocates an IP and populates the effective_ip_address + # field. + "ipVersion": "A String", # Required. The IP version of the IP. It is IPv4 or IPv6. + "subnetworkRangeNames": [ # Input only. Defines candidate secondary ranges for allocating the alias IPs. If it is + # empty, the IP would be allocated from the primary range of the + # subnetwork. + # This field is INPUT_ONLY and used only during resource creation, + # hence a repeated field is acceptable as per AIP-2825. It is not mutable. + "A String", + ], + }, + }, + "annotations": { # Optional. The annotations map for the AliasNetworkEndpoint. + # Keys and values must follow Kubernetes limits. + # Keys must be a maximum of 253 characters. Keys should be dot-namespaced, + # following the Kubernetes annotation key format: + # `([DNS subdomain]/)?[RFC 1123 name]`. + # Values must be a maximum of 256 characters. + "a_key": "A String", + }, + "creationTimestamp": "A String", # Output only. [Output Only] Creation timestamp in RFC3339 text format. + "description": "A String", # Optional. An optional description of this resource. Provide this field when you + # create the resource. + "host": { # Information about the host instance. # Required. The host that the alias network endpoint is bound to. + "instance": "A String", # Required. The URL to the instance that this alias network endpoint is bound to. + }, + "id": "A String", # Output only. [Output Only] The unique identifier for the resource. This identifier is + # defined by the server. + "labels": { # Optional. The labels map for the AliasNetworkEndpoint. + "a_key": "A String", + }, + "name": "A String", # Identifier. Name of the resource. Provided by the client when the resource is created. + # The name must be 1-63 characters long, and comply with RFC1035. + "securityTagInheritance": "A String", # Optional. The security tag inheritance setting for the AliasNetworkEndpoint. + "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "status": { # Represents the current status of the AliasNetworkEndpoint. # Output only. [Output Only] Current status of the alias network endpoint. + "network": "A String", # Output only. [Output Only] The URL of the network that this alias network endpoint + # belongs to. + "state": "A String", # Output only. [Output Only] The state of the alias network endpoint. + }, + "subnetwork": "A String", # Required. The URL of the subnetwork that this alias network endpoint belongs to. + "updateTime": "A String", # Output only. [Output Only] Update timestamp in RFC3339 text format. + }, + ], + "nextPageToken": "A String", # [Output Only] This token allows you to get the next page of results for + # list requests. If the number of results is larger thanmaxResults, use the nextPageToken as a value for + # the query parameter pageToken in the next list request. + # Subsequent list requests will have their own nextPageToken to + # continue paging through the results. + "selfLink": "A String", # [Output Only] Server-defined URL for this resource. + "unreachables": [ # Output only. [Output Only] Unreachable resources. + "A String", + ], + "warning": { # [Output Only] Informational warning message. + "code": "A String", # [Output Only] A warning code, if applicable. For example, Compute + # Engine returns NO_RESULTS_ON_PAGE if there + # are no results in the response. + "data": [ # [Output Only] Metadata about this warning in key: + # value format. For example: + # + # "data": [ + # { + # "key": "scope", + # "value": "zones/us-east1-d" + # }] + { + "key": "A String", # [Output Only] A key that provides more detail on the warning being + # returned. For example, for warnings where there are no results in a list + # request for a particular zone, this key might be scope and + # the key value might be the zone name. Other examples might be a key + # indicating a deprecated resource and a suggested replacement, or a + # warning about invalid network settings (for example, if an instance + # attempts to perform IP forwarding but is not enabled for IP forwarding). + "value": "A String", # [Output Only] A warning data value corresponding to the key. + }, + ], + "message": "A String", # [Output Only] A human-readable description of the warning code. + }, +}+++ + \ No newline at end of file diff --git a/docs/dyn/compute_preview.globalAddresses.html b/docs/dyn/compute_preview.globalAddresses.html index 9a401ed24e..8e71e4ac69 100644 --- a/docs/dyn/compute_preview.globalAddresses.html +++ b/docs/dyn/compute_preview.globalAddresses.html @@ -505,6 +505,9 @@list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -548,6 +551,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and @@ -649,6 +655,9 @@Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -692,6 +701,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and @@ -1175,6 +1187,9 @@Method Details
# character must be a lowercase letter or digit. "network": "A String", # The URL of the network in which to reserve the address. This field can # only be used with INTERNAL type with theVPC_PEERING purpose. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this address comes from in the + # following format: + # projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkTier": "A String", # This signifies the networking tier used for configuring this address and # can only take the following values: PREMIUM orSTANDARD. Internal IP addresses are always Premium Tier; # global external IP addresses are always Premium Tier; regional external IP @@ -1218,6 +1233,9 @@Method Details
# the HTTP request URL. *This field is not applicable to global # addresses.* "selfLink": "A String", # [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. Producer Service's Service class ID for the region of this address. Can + # only be used with network_attachment. It is not possible to use on its own; + # however, network_attachment can be used without service_class_id. "status": "A String", # Output only. [Output Only] The status of the address, which can be one ofRESERVING, RESERVED, or IN_USE. # An address that is RESERVING is currently in the process of # being reserved. A RESERVED address is currently reserved and diff --git a/docs/dyn/compute_preview.globalNetworkEndpointGroups.html b/docs/dyn/compute_preview.globalNetworkEndpointGroups.html index 2689bfe0a0..06680c6e82 100644 --- a/docs/dyn/compute_preview.globalNetworkEndpointGroups.html +++ b/docs/dyn/compute_preview.globalNetworkEndpointGroups.html @@ -1267,6 +1267,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -1288,6 +1290,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -1452,6 +1458,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -1473,6 +1481,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -2006,6 +2018,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -2027,6 +2041,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. diff --git a/docs/dyn/compute_preview.haControllers.html b/docs/dyn/compute_preview.haControllers.html index c05c512d62..6a6e80328e 100644 --- a/docs/dyn/compute_preview.haControllers.html +++ b/docs/dyn/compute_preview.haControllers.html @@ -255,6 +255,12 @@Method Details
# naming consistent with NetworkInterface configuration on Instances. }, }, + "params": { # Additional HA Controller params. # Input only. Additional parameters passed with the request, but not persisted as part + # of the resource. Both `params: null` and `params: {}` are treated + # identically as a no-op. + "migrateDisksToRegional": True or False, # Specifies whether to automatically migrate all disks of an instance from + # zonal to regional before attempting to create the HaController. + }, "region": "A String", # Output only. [Output Only] URL of the region where the resource resides. # You must specify this field as part of the HTTP request URL. # It is not settable as a field in the request body. @@ -617,6 +623,35 @@Method Details
# Value: ZoneConfiguration "a_key": { # Config for a zone that the HaController may use for running the VM # instance. + "instanceNetworkInterfaces": { # Optional. The network interface configuration of a VM instance in this zone, + # keyed by the name of the network interface. A VM network interface + # uses the nicN naming format, where N is a + # value between 0 and 7. The default interface + # is nic0. Use this field to choose the internal IP + # addresses of the VM instance in this zone instead of letting Compute + # Engine assign them automatically. + # + # + # - Each key must be a name of a network interface that exists on + # the VM instance. Otherwise, the request fails. + # - Network interfaces that are omitted are assigned an ephemeral + # internal IP address. + # - Every zone must configure the same network interfaces and the + # same IP address families. + # - For the zone that currently holds the VM instance, this field + # is populated automatically, and any addresses you specify must + # match the instance's current addresses. + # - Only internal IP addresses are supported, and this field can + # only be set when the HA Controller is created. + "a_key": { # The network interface configuration of the VM instance. + "ipv6Address": "A String", # Optional. Optional IPv6 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + "networkIP": "A String", # Optional. Optional IPv4 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + }, + }, "nodeAffinities": [ # A set of node affinity configurations. Refer toConfiguring node # affinity for more information. # Overrides reservationAffinity. @@ -1401,6 +1436,12 @@Method Details
# naming consistent with NetworkInterface configuration on Instances. }, }, + "params": { # Additional HA Controller params. # Input only. Additional parameters passed with the request, but not persisted as part + # of the resource. Both `params: null` and `params: {}` are treated + # identically as a no-op. + "migrateDisksToRegional": True or False, # Specifies whether to automatically migrate all disks of an instance from + # zonal to regional before attempting to create the HaController. + }, "region": "A String", # Output only. [Output Only] URL of the region where the resource resides. # You must specify this field as part of the HTTP request URL. # It is not settable as a field in the request body. @@ -1763,6 +1804,35 @@Method Details
# Value: ZoneConfiguration "a_key": { # Config for a zone that the HaController may use for running the VM # instance. + "instanceNetworkInterfaces": { # Optional. The network interface configuration of a VM instance in this zone, + # keyed by the name of the network interface. A VM network interface + # uses the nicN naming format, where N is a + # value between 0 and 7. The default interface + # is nic0. Use this field to choose the internal IP + # addresses of the VM instance in this zone instead of letting Compute + # Engine assign them automatically. + # + # + # - Each key must be a name of a network interface that exists on + # the VM instance. Otherwise, the request fails. + # - Network interfaces that are omitted are assigned an ephemeral + # internal IP address. + # - Every zone must configure the same network interfaces and the + # same IP address families. + # - For the zone that currently holds the VM instance, this field + # is populated automatically, and any addresses you specify must + # match the instance's current addresses. + # - Only internal IP addresses are supported, and this field can + # only be set when the HA Controller is created. + "a_key": { # The network interface configuration of the VM instance. + "ipv6Address": "A String", # Optional. Optional IPv6 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + "networkIP": "A String", # Optional. Optional IPv4 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + }, + }, "nodeAffinities": [ # A set of node affinity configurations. Refer toConfiguring node # affinity for more information. # Overrides reservationAffinity. @@ -1851,6 +1921,12 @@Method Details
# naming consistent with NetworkInterface configuration on Instances. }, }, + "params": { # Additional HA Controller params. # Input only. Additional parameters passed with the request, but not persisted as part + # of the resource. Both `params: null` and `params: {}` are treated + # identically as a no-op. + "migrateDisksToRegional": True or False, # Specifies whether to automatically migrate all disks of an instance from + # zonal to regional before attempting to create the HaController. + }, "region": "A String", # Output only. [Output Only] URL of the region where the resource resides. # You must specify this field as part of the HTTP request URL. # It is not settable as a field in the request body. @@ -2213,6 +2289,35 @@Method Details
# Value: ZoneConfiguration "a_key": { # Config for a zone that the HaController may use for running the VM # instance. + "instanceNetworkInterfaces": { # Optional. The network interface configuration of a VM instance in this zone, + # keyed by the name of the network interface. A VM network interface + # uses the nicN naming format, where N is a + # value between 0 and 7. The default interface + # is nic0. Use this field to choose the internal IP + # addresses of the VM instance in this zone instead of letting Compute + # Engine assign them automatically. + # + # + # - Each key must be a name of a network interface that exists on + # the VM instance. Otherwise, the request fails. + # - Network interfaces that are omitted are assigned an ephemeral + # internal IP address. + # - Every zone must configure the same network interfaces and the + # same IP address families. + # - For the zone that currently holds the VM instance, this field + # is populated automatically, and any addresses you specify must + # match the instance's current addresses. + # - Only internal IP addresses are supported, and this field can + # only be set when the HA Controller is created. + "a_key": { # The network interface configuration of the VM instance. + "ipv6Address": "A String", # Optional. Optional IPv6 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + "networkIP": "A String", # Optional. Optional IPv4 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + }, + }, "nodeAffinities": [ # A set of node affinity configurations. Refer toConfiguring node # affinity for more information. # Overrides reservationAffinity. @@ -2675,6 +2780,12 @@Method Details
# naming consistent with NetworkInterface configuration on Instances. }, }, + "params": { # Additional HA Controller params. # Input only. Additional parameters passed with the request, but not persisted as part + # of the resource. Both `params: null` and `params: {}` are treated + # identically as a no-op. + "migrateDisksToRegional": True or False, # Specifies whether to automatically migrate all disks of an instance from + # zonal to regional before attempting to create the HaController. + }, "region": "A String", # Output only. [Output Only] URL of the region where the resource resides. # You must specify this field as part of the HTTP request URL. # It is not settable as a field in the request body. @@ -3037,6 +3148,35 @@Method Details
# Value: ZoneConfiguration "a_key": { # Config for a zone that the HaController may use for running the VM # instance. + "instanceNetworkInterfaces": { # Optional. The network interface configuration of a VM instance in this zone, + # keyed by the name of the network interface. A VM network interface + # uses the nicN naming format, where N is a + # value between 0 and 7. The default interface + # is nic0. Use this field to choose the internal IP + # addresses of the VM instance in this zone instead of letting Compute + # Engine assign them automatically. + # + # + # - Each key must be a name of a network interface that exists on + # the VM instance. Otherwise, the request fails. + # - Network interfaces that are omitted are assigned an ephemeral + # internal IP address. + # - Every zone must configure the same network interfaces and the + # same IP address families. + # - For the zone that currently holds the VM instance, this field + # is populated automatically, and any addresses you specify must + # match the instance's current addresses. + # - Only internal IP addresses are supported, and this field can + # only be set when the HA Controller is created. + "a_key": { # The network interface configuration of the VM instance. + "ipv6Address": "A String", # Optional. Optional IPv6 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + "networkIP": "A String", # Optional. Optional IPv4 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + }, + }, "nodeAffinities": [ # A set of node affinity configurations. Refer toConfiguring node # affinity for more information. # Overrides reservationAffinity. @@ -3175,6 +3315,12 @@Method Details
# naming consistent with NetworkInterface configuration on Instances. }, }, + "params": { # Additional HA Controller params. # Input only. Additional parameters passed with the request, but not persisted as part + # of the resource. Both `params: null` and `params: {}` are treated + # identically as a no-op. + "migrateDisksToRegional": True or False, # Specifies whether to automatically migrate all disks of an instance from + # zonal to regional before attempting to create the HaController. + }, "region": "A String", # Output only. [Output Only] URL of the region where the resource resides. # You must specify this field as part of the HTTP request URL. # It is not settable as a field in the request body. @@ -3537,6 +3683,35 @@Method Details
# Value: ZoneConfiguration "a_key": { # Config for a zone that the HaController may use for running the VM # instance. + "instanceNetworkInterfaces": { # Optional. The network interface configuration of a VM instance in this zone, + # keyed by the name of the network interface. A VM network interface + # uses the nicN naming format, where N is a + # value between 0 and 7. The default interface + # is nic0. Use this field to choose the internal IP + # addresses of the VM instance in this zone instead of letting Compute + # Engine assign them automatically. + # + # + # - Each key must be a name of a network interface that exists on + # the VM instance. Otherwise, the request fails. + # - Network interfaces that are omitted are assigned an ephemeral + # internal IP address. + # - Every zone must configure the same network interfaces and the + # same IP address families. + # - For the zone that currently holds the VM instance, this field + # is populated automatically, and any addresses you specify must + # match the instance's current addresses. + # - Only internal IP addresses are supported, and this field can + # only be set when the HA Controller is created. + "a_key": { # The network interface configuration of the VM instance. + "ipv6Address": "A String", # Optional. Optional IPv6 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + "networkIP": "A String", # Optional. Optional IPv4 address to assign to the instance's network interface. + # If omitted, an ephemeral IP may be assigned depending on the instance's + # network interface stack type. + }, + }, "nodeAffinities": [ # A set of node affinity configurations. Refer toConfiguring node # affinity for more information. # Overrides reservationAffinity. diff --git a/docs/dyn/compute_preview.html b/docs/dyn/compute_preview.html index 089b55848f..535d3e1f85 100644 --- a/docs/dyn/compute_preview.html +++ b/docs/dyn/compute_preview.html @@ -99,6 +99,11 @@Instance Methods
Returns the advice Resource.
+ +Returns the aliasNetworkEndpoints Resource.
+ diff --git a/docs/dyn/compute_preview.instanceTemplates.html b/docs/dyn/compute_preview.instanceTemplates.html index 52063d5847..89a9d507b8 100644 --- a/docs/dyn/compute_preview.instanceTemplates.html +++ b/docs/dyn/compute_preview.instanceTemplates.html @@ -826,6 +826,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -906,6 +921,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1148,6 +1178,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -2256,6 +2289,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2336,6 +2384,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2578,6 +2641,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -3663,6 +3729,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3743,6 +3824,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3985,6 +4081,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -5082,6 +5181,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -5162,6 +5276,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -5404,6 +5533,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical diff --git a/docs/dyn/compute_preview.instances.html b/docs/dyn/compute_preview.instances.html index 25e63df214..aac3531dff 100644 --- a/docs/dyn/compute_preview.instances.html +++ b/docs/dyn/compute_preview.instances.html @@ -265,6 +265,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -625,6 +640,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -705,6 +735,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2231,6 +2276,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2311,6 +2371,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2513,6 +2588,9 @@Method Details
# specified in the spread placement policy attached to the instance. "gracefulShutdownTimestamp": "A String", # Output only. Specifies the timestamp, when the instance will start graceful shutdown # process, in RFC3339 text format. + "protectionTier": "A String", # Output only. Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "terminationTimestamp": "A String", # Time in future when the instance will be terminated inRFC3339 text format. }, "shutdownDetails": { # Specifies if the instance is in `PENDING_STOP` state or there is a # Output only. [Output Only] Details about the instance stopping state. @@ -2661,6 +2739,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -4734,6 +4815,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -4814,6 +4910,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -5056,6 +5167,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -7439,6 +7553,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -7519,6 +7648,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -7721,6 +7865,9 @@Method Details
# specified in the spread placement policy attached to the instance. "gracefulShutdownTimestamp": "A String", # Output only. Specifies the timestamp, when the instance will start graceful shutdown # process, in RFC3339 text format. + "protectionTier": "A String", # Output only. Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "terminationTimestamp": "A String", # Time in future when the instance will be terminated inRFC3339 text format. }, "shutdownDetails": { # Specifies if the instance is in `PENDING_STOP` state or there is a # Output only. [Output Only] Details about the instance stopping state. @@ -7869,6 +8016,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -9806,6 +9956,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -9886,6 +10051,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -10088,6 +10268,9 @@Method Details
# specified in the spread placement policy attached to the instance. "gracefulShutdownTimestamp": "A String", # Output only. Specifies the timestamp, when the instance will start graceful shutdown # process, in RFC3339 text format. + "protectionTier": "A String", # Output only. Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "terminationTimestamp": "A String", # Time in future when the instance will be terminated inRFC3339 text format. }, "shutdownDetails": { # Specifies if the instance is in `PENDING_STOP` state or there is a # Output only. [Output Only] Details about the instance stopping state. @@ -10236,6 +10419,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -11474,6 +11660,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -11554,6 +11755,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -11756,6 +11972,9 @@Method Details
# specified in the spread placement policy attached to the instance. "gracefulShutdownTimestamp": "A String", # Output only. Specifies the timestamp, when the instance will start graceful shutdown # process, in RFC3339 text format. + "protectionTier": "A String", # Output only. Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "terminationTimestamp": "A String", # Time in future when the instance will be terminated inRFC3339 text format. }, "shutdownDetails": { # Specifies if the instance is in `PENDING_STOP` state or there is a # Output only. [Output Only] Details about the instance stopping state. @@ -11904,6 +12123,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -17376,6 +17598,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -21672,6 +21897,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -21752,6 +21992,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -21954,6 +22209,9 @@Method Details
# specified in the spread placement policy attached to the instance. "gracefulShutdownTimestamp": "A String", # Output only. Specifies the timestamp, when the instance will start graceful shutdown # process, in RFC3339 text format. + "protectionTier": "A String", # Output only. Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "terminationTimestamp": "A String", # Time in future when the instance will be terminated inRFC3339 text format. }, "shutdownDetails": { # Specifies if the instance is in `PENDING_STOP` state or there is a # Output only. [Output Only] Details about the instance stopping state. @@ -22102,6 +22360,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -22573,6 +22834,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -23255,6 +23531,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -23335,6 +23626,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but diff --git a/docs/dyn/compute_preview.machineImages.html b/docs/dyn/compute_preview.machineImages.html index c0dd3942c8..835a21f76b 100644 --- a/docs/dyn/compute_preview.machineImages.html +++ b/docs/dyn/compute_preview.machineImages.html @@ -1023,6 +1023,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1103,6 +1118,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1345,6 +1375,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -1789,6 +1822,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1869,6 +1917,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2078,6 +2141,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -3095,6 +3161,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3175,6 +3256,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3417,6 +3513,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -3861,6 +3960,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3941,6 +4055,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -4150,6 +4279,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -5180,6 +5312,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -5260,6 +5407,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -5502,6 +5664,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -5946,6 +6111,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -6026,6 +6206,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -6235,6 +6430,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical diff --git a/docs/dyn/compute_preview.networkEndpointGroups.html b/docs/dyn/compute_preview.networkEndpointGroups.html index bbc4f89a8b..443ec763b6 100644 --- a/docs/dyn/compute_preview.networkEndpointGroups.html +++ b/docs/dyn/compute_preview.networkEndpointGroups.html @@ -355,6 +355,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -376,6 +378,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -1637,6 +1643,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -1658,6 +1666,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -1825,6 +1837,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -1846,6 +1860,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -2381,6 +2399,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -2402,6 +2422,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. diff --git a/docs/dyn/compute_preview.regionInstanceTemplates.html b/docs/dyn/compute_preview.regionInstanceTemplates.html index 5f2c3e5552..b6bf625370 100644 --- a/docs/dyn/compute_preview.regionInstanceTemplates.html +++ b/docs/dyn/compute_preview.regionInstanceTemplates.html @@ -1030,6 +1030,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1110,6 +1125,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1352,6 +1382,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -2067,6 +2100,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2147,6 +2195,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -2389,6 +2452,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical @@ -3487,6 +3553,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3567,6 +3648,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -3809,6 +3905,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical diff --git a/docs/dyn/compute_preview.regionInstances.html b/docs/dyn/compute_preview.regionInstances.html index 5b282720e1..9bfb75e898 100644 --- a/docs/dyn/compute_preview.regionInstances.html +++ b/docs/dyn/compute_preview.regionInstances.html @@ -1169,6 +1169,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1249,6 +1264,21 @@Method Details
# assign an external IPv6 address from the instance's subnetwork. "externalIpv6PrefixLength": 42, # Applies to ipv6AccessConfigs only. The prefix length of the # external IPv6 range. + "ipCollection": "A String", # Reference to the source of external IPv4 addresses, + # like a PublicDelegatedPrefix (PDP) for BYOIP. + # The PDP must support enhanced IPv4 allocations. + # + # Use one of the following formats to specify a PDP when allocating an + # external IPv4 address using BYOIP. + # + # - + # Full resource URL, as inhttps://www.googleapis.com/compute/v1/projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - + # Partial URL, as in + # + # + # - projects/projectId/regions/region/publicDelegatedPrefixes/pdp-name + # - regions/region/publicDelegatedPrefixes/pdp-name "kind": "compute#accessConfig", # Output only. [Output Only] Type of the resource. Alwayscompute#accessConfig for access configs. "name": "A String", # The name of this access configuration. In accessConfigs # (IPv4), the default and recommended name is External NAT, but @@ -1491,6 +1521,9 @@Method Details
# to 315,576,000,000 inclusive. Note: these bounds are computed from: # 60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years }, + "protectionTier": "A String", # Protection tier for the workload which specifies the workload + # expectations in the event of infrastructure failures at data center + # (e.g. power and/or cooling failures). "provisioningModel": "A String", # Specifies the provisioning model of the instance. "skipGuestOsShutdown": True or False, # Default is false and there will be 120 seconds between GCE ACPI G2 Soft # Off and ACPI G3 Mechanical diff --git a/docs/dyn/compute_preview.regionNetworkEndpointGroups.html b/docs/dyn/compute_preview.regionNetworkEndpointGroups.html index 0a71134257..dd43fac62c 100644 --- a/docs/dyn/compute_preview.regionNetworkEndpointGroups.html +++ b/docs/dyn/compute_preview.regionNetworkEndpointGroups.html @@ -1278,6 +1278,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -1299,6 +1301,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -1466,6 +1472,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -1487,6 +1495,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. @@ -2022,6 +2034,8 @@Method Details
# This field cannot be set for NEGs with networkEndpointType set toSERVERLESS or PRIVATE_SERVICE_CONNECT and for # global NEGs. # For all other network endpoint types, this field is required. + "networkAttachment": "A String", # Optional. The URL of the network attachment that this resource + # belongs to.projects/{project}/regions/{region_name}/networkAttachments/{network_attachment_name}. "networkEndpointType": "A String", # Type of network endpoints in this network endpoint group. Can be one ofGCE_VM_IP, GCE_VM_IP_PORT,NON_GCP_PRIVATE_IP_PORT, INTERNET_FQDN_PORT,INTERNET_IP_PORT, SERVERLESS,PRIVATE_SERVICE_CONNECT, GCE_VM_IP_PORTMAP. "pscData": { # All data that is specifically relevant to only network endpoint groups of # Optional. Only valid when networkEndpointType isPRIVATE_SERVICE_CONNECT. # type PRIVATE_SERVICE_CONNECT. @@ -2043,6 +2057,10 @@Method Details
"region": "A String", # Output only. [Output Only] The URL of theregion # where the network endpoint group is located. "selfLink": "A String", # Output only. [Output Only] Server-defined URL for the resource. + "serviceClassId": "A String", # Optional. The service class ID associated with this resource. + # Can only be used with network_attachment. It is not possible to + # use on its own; however, network_attachment can be used without + # service_class_id. "size": 42, # Output only. [Output only] Number of network endpoints in the network endpoint group. "subnetwork": "A String", # Optional URL of the subnetwork to which all network endpoints in the NEG # belong. diff --git a/docs/dyn/dataplex_v1.projects.locations.dataScans.html b/docs/dyn/dataplex_v1.projects.locations.dataScans.html index 3ccef8877b..b99154873e 100644 --- a/docs/dyn/dataplex_v1.projects.locations.dataScans.html +++ b/docs/dyn/dataplex_v1.projects.locations.dataScans.html @@ -178,10 +178,6 @@Method Details
"encoding": "A String", # Optional. The character encoding of the data. The default is UTF-8. "typeInferenceDisabled": True or False, # Optional. Whether to disable the inference of data types for JSON data. If true, all columns are registered as their primitive types (strings, number, or boolean). }, - "unstructuredDataOptions": { # Describes options for unstructured data discovery. # Optional. Specifies configuration for unstructured data discovery. - "globalEndpointEnabled": True or False, # Optional. Whether to use the global model endpoint. - "semanticInferenceEnabled": True or False, # Optional. Specifies whether deeper semantic inference over the objects' contents using GenAI is enabled. - }, }, }, "dataDocumentationResult": { # The output of a DataDocumentation scan. # Output only. The result of a data documentation scan. @@ -654,85 +650,6 @@Method Details
"state": "A String", # Output only. Current state of the DataScan. "type": "A String", # Output only. The type of DataScan. "uid": "A String", # Output only. System generated globally unique ID for the scan. This ID will be different if the scan is deleted and re-created with the same name. - "unstructuredDataProfileResult": { # Contains the result of an unstructured data profile scan. # Output only. The result of an unstructured data profile scan. - "description": "A String", # Output only. The inferred description. - "graphProfile": { # Contains the strict structure for graph-profile for semantic inference scan result. # Output only. The inferred graph profile. - "edgeTypes": [ # Output only. Edge types. - { # Represents a type of edge (relationship) in the graph. - "description": "A String", # Output only. Description of the edge type. - "extractionHints": { # Extraction hints (edge-level). # Output only. Extraction hints for the edge. - "cardinality": "A String", # Output only. Expected connectivity topology and bounds of this relationship. Format: "Topology - Description" Example: "1:N - One company can have multiple financial reports." - }, - "fields": [ # Output only. Fields of the edge type. - { # Represents a field in a node or edge type. - "dataType": "A String", # Output only. The data type of the field, e.g., STRING, INTEGER, DATE. - "description": "A String", # Output only. Description of the field. - "extractionHints": { # Extraction hints (field-level). # Output only. Extraction hints for the field. - "normalization": "A String", # Output only. Standardizes extracted data (e.g., to ISO 3166-1 alpha-2). - "synthesis": "A String", # Output only. Generates value from other data instead of direct extraction (e.g., hashing). - }, - "fields": [ # Output only. Sub-fields of this field (for STRUCT types). - # Object with schema name: GoogleCloudDataplexV1GraphProfileField - ], - "metadataType": "A String", # Output only. The mapped metadata type. - "mode": "A String", # Output only. The mode of the field. - "name": "A String", # Output only. Name of the field. - }, - ], - "foreignKeys": [ # Output only. Defines the Foreign Key constraints for the edge. - { # Represents a foreign key constraint. - "description": "A String", # Output only. Description of the foreign key. - "fieldMappings": [ # Output only. Field Mappings. Mappings between local fields and the fields they reference in the referenced node type. - { # Maps a local field to a referenced field. - "field": "A String", # Output only. Local field name forming part of the foreign key. - "referencedField": "A String", # Output only. Field name in the referenced node type. - }, - ], - "name": "A String", # Output only. Name of the foreign key constraint. - "referencedNodeType": "A String", # Output only. The node type this constraint references. - }, - ], - "name": "A String", # Output only. Name of the edge type. - "sourceNodeType": "A String", # Output only. Source node type. - "targetNodeType": "A String", # Output only. Target node type. - }, - ], - "nodeTypes": [ # Output only. Node types. - { # Represents a type of node in the graph. - "description": "A String", # Output only. Description of the node type. - "extractionHints": { # Extraction hints (node-level). # Output only. Extraction hints for the node. - "cardinality": "A String", # Output only. Expected occurrence frequency of this node type within a document. Format: "Bounds - Description" Example: "0:N - A document may contain multiple people names." - }, - "fields": [ # Output only. Fields of the node type. - { # Represents a field in a node or edge type. - "dataType": "A String", # Output only. The data type of the field, e.g., STRING, INTEGER, DATE. - "description": "A String", # Output only. Description of the field. - "extractionHints": { # Extraction hints (field-level). # Output only. Extraction hints for the field. - "normalization": "A String", # Output only. Standardizes extracted data (e.g., to ISO 3166-1 alpha-2). - "synthesis": "A String", # Output only. Generates value from other data instead of direct extraction (e.g., hashing). - }, - "fields": [ # Output only. Sub-fields of this field (for STRUCT types). - # Object with schema name: GoogleCloudDataplexV1GraphProfileField - ], - "metadataType": "A String", # Output only. The mapped metadata type. - "mode": "A String", # Output only. The mode of the field. - "name": "A String", # Output only. Name of the field. - }, - ], - "name": "A String", # Output only. Name of the node type. - "primaryKeys": [ # Output only. Field names forming the primary keys. The order in this array defines the key's ordinal positions for composite keys. - "A String", - ], - }, - ], - }, - "partialFailureMessage": "A String", # Output only. Optional message for partial failures (e.g. node type extraction failed). - }, - "unstructuredDataProfileSpec": { # Contains the specification for an unstructured data profile scan. # Optional. Settings for an unstructured data profile scan. - "customizedPrompt": "A String", # Optional. Customized prompt for unstructured data profile. The field will be used as part of the prompt, could be some instruction, specifying skill, or specific area to focus. - "globalEndpointEnabled": True or False, # Optional. Whether to use the global model. - "graphProfilePublishingEnabled": True or False, # Optional. Whether to publish graph-profile as aspect on the catalog entry. - }, "updateTime": "A String", # Output only. The time when the scan was last updated. } @@ -990,10 +907,6 @@Method Details
"encoding": "A String", # Optional. The character encoding of the data. The default is UTF-8. "typeInferenceDisabled": True or False, # Optional. Whether to disable the inference of data types for JSON data. If true, all columns are registered as their primitive types (strings, number, or boolean). }, - "unstructuredDataOptions": { # Describes options for unstructured data discovery. # Optional. Specifies configuration for unstructured data discovery. - "globalEndpointEnabled": True or False, # Optional. Whether to use the global model endpoint. - "semanticInferenceEnabled": True or False, # Optional. Specifies whether deeper semantic inference over the objects' contents using GenAI is enabled. - }, }, }, "dataDocumentationResult": { # The output of a DataDocumentation scan. # Output only. The result of a data documentation scan. @@ -1466,85 +1379,6 @@Method Details
"state": "A String", # Output only. Current state of the DataScan. "type": "A String", # Output only. The type of DataScan. "uid": "A String", # Output only. System generated globally unique ID for the scan. This ID will be different if the scan is deleted and re-created with the same name. - "unstructuredDataProfileResult": { # Contains the result of an unstructured data profile scan. # Output only. The result of an unstructured data profile scan. - "description": "A String", # Output only. The inferred description. - "graphProfile": { # Contains the strict structure for graph-profile for semantic inference scan result. # Output only. The inferred graph profile. - "edgeTypes": [ # Output only. Edge types. - { # Represents a type of edge (relationship) in the graph. - "description": "A String", # Output only. Description of the edge type. - "extractionHints": { # Extraction hints (edge-level). # Output only. Extraction hints for the edge. - "cardinality": "A String", # Output only. Expected connectivity topology and bounds of this relationship. Format: "Topology - Description" Example: "1:N - One company can have multiple financial reports." - }, - "fields": [ # Output only. Fields of the edge type. - { # Represents a field in a node or edge type. - "dataType": "A String", # Output only. The data type of the field, e.g., STRING, INTEGER, DATE. - "description": "A String", # Output only. Description of the field. - "extractionHints": { # Extraction hints (field-level). # Output only. Extraction hints for the field. - "normalization": "A String", # Output only. Standardizes extracted data (e.g., to ISO 3166-1 alpha-2). - "synthesis": "A String", # Output only. Generates value from other data instead of direct extraction (e.g., hashing). - }, - "fields": [ # Output only. Sub-fields of this field (for STRUCT types). - # Object with schema name: GoogleCloudDataplexV1GraphProfileField - ], - "metadataType": "A String", # Output only. The mapped metadata type. - "mode": "A String", # Output only. The mode of the field. - "name": "A String", # Output only. Name of the field. - }, - ], - "foreignKeys": [ # Output only. Defines the Foreign Key constraints for the edge. - { # Represents a foreign key constraint. - "description": "A String", # Output only. Description of the foreign key. - "fieldMappings": [ # Output only. Field Mappings. Mappings between local fields and the fields they reference in the referenced node type. - { # Maps a local field to a referenced field. - "field": "A String", # Output only. Local field name forming part of the foreign key. - "referencedField": "A String", # Output only. Field name in the referenced node type. - }, - ], - "name": "A String", # Output only. Name of the foreign key constraint. - "referencedNodeType": "A String", # Output only. The node type this constraint references. - }, - ], - "name": "A String", # Output only. Name of the edge type. - "sourceNodeType": "A String", # Output only. Source node type. - "targetNodeType": "A String", # Output only. Target node type. - }, - ], - "nodeTypes": [ # Output only. Node types. - { # Represents a type of node in the graph. - "description": "A String", # Output only. Description of the node type. - "extractionHints": { # Extraction hints (node-level). # Output only. Extraction hints for the node. - "cardinality": "A String", # Output only. Expected occurrence frequency of this node type within a document. Format: "Bounds - Description" Example: "0:N - A document may contain multiple people names." - }, - "fields": [ # Output only. Fields of the node type. - { # Represents a field in a node or edge type. - "dataType": "A String", # Output only. The data type of the field, e.g., STRING, INTEGER, DATE. - "description": "A String", # Output only. Description of the field. - "extractionHints": { # Extraction hints (field-level). # Output only. Extraction hints for the field. - "normalization": "A String", # Output only. Standardizes extracted data (e.g., to ISO 3166-1 alpha-2). - "synthesis": "A String", # Output only. Generates value from other data instead of direct extraction (e.g., hashing). - }, - "fields": [ # Output only. Sub-fields of this field (for STRUCT types). - # Object with schema name: GoogleCloudDataplexV1GraphProfileField - ], - "metadataType": "A String", # Output only. The mapped metadata type. - "mode": "A String", # Output only. The mode of the field. - "name": "A String", # Output only. Name of the field. - }, - ], - "name": "A String", # Output only. Name of the node type. - "primaryKeys": [ # Output only. Field names forming the primary keys. The order in this array defines the key's ordinal positions for composite keys. - "A String", - ], - }, - ], - }, - "partialFailureMessage": "A String", # Output only. Optional message for partial failures (e.g. node type extraction failed). - }, - "unstructuredDataProfileSpec": { # Contains the specification for an unstructured data profile scan. # Optional. Settings for an unstructured data profile scan. - "customizedPrompt": "A String", # Optional. Customized prompt for unstructured data profile. The field will be used as part of the prompt, could be some instruction, specifying skill, or specific area to focus. - "globalEndpointEnabled": True or False, # Optional. Whether to use the global model. - "graphProfilePublishingEnabled": True or False, # Optional. Whether to publish graph-profile as aspect on the catalog entry. - }, "updateTime": "A String", # Output only. The time when the scan was last updated. }
+ batchGet(parent, names=None, x__xgafv=None)
Retrieves multiple automation sessions in a single batch request. Sessions are returned in the same order as the `names` in the request. If any of the requested sessions does not exist, the whole request fails with `NOT_FOUND` and no sessions are returned. When calling this method over HTTP/REST with a large number of `names`, the request URL may exceed the maximum URL length (about 16 KB) and be rejected. In that case, send a `POST` request to the same URL with the `X-HTTP-Method-Override: GET` header and pass the request parameters in a JSON (`application/json`) or form-encoded (`application/x-www-form-urlencoded`) body. See https://cloud.google.com/apis/docs/http#long_request_urls.
cancel(name, body=None, x__xgafv=None)
Cancels an in-progress automation session. This RPC returns immediately and cancellation proceeds asynchronously. If the session is already finished, this RPC will have no effect.
@@ -96,6 +99,363 @@list_next(previous_request, previous_response)
Retrieves the next page of results.
batchGet(parent, names=None, x__xgafv=None)
+ Retrieves multiple automation sessions in a single batch request. Sessions are returned in the same order as the `names` in the request. If any of the requested sessions does not exist, the whole request fails with `NOT_FOUND` and no sessions are returned. When calling this method over HTTP/REST with a large number of `names`, the request URL may exceed the maximum URL length (about 16 KB) and be rejected. In that case, send a `POST` request to the same URL with the `X-HTTP-Method-Override: GET` header and pass the request parameters in a JSON (`application/json`) or form-encoded (`application/x-www-form-urlencoded`) body. See https://cloud.google.com/apis/docs/http#long_request_urls.
+
+Args:
+ parent: string, Required. The parent resource shared by all sessions being retrieved. Format: `projects/{project}/locations/{location}`. The `parent` field in the `BatchGetSessionsRequest` message must match the `parent` of all `Session` resource names in `names`. (required)
+ names: string, Required. The names of the sessions to retrieve. A maximum of 500 sessions can be retrieved in a batch. Format: `projects/{project}/locations/{location}/sessions/{session}`. (repeated)
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Response including the requested sessions.
+ "sessions": [ # The requested sessions, in the same order as the `names` in the request.
+ { # A session resource in the AutomationSession API. At a high level, `Session` describes the configuration of one or multiple jobs, the state transitions it goes through, and the results.
+ "name": "A String", # Identifier. The resource name of the session. Format: `projects/{project}/locations/{location}/sessions/{session}`.
+ "sessionConfig": { # SessionConfig is used to create a session. # Required. Configuration used to create the session.
+ "displayName": "A String", # Optional. User-settable, human-readable name for the session. Maximum size is 63 bytes when encoded as UTF-8. If set, must match regex: `^A-Za-z0-9*$`.
+ "jobConfigs": [ # Required. Configs of the jobs in the session.
+ { # The configuration of a job.
+ "action": { # The action to be performed in a job. # Required. Job action.
+ "androidInstrumentationTest": { # The configuration of an Android instrumentation test. See https://developer.android.com/training/testing/instrumented-tests for more information on Android instrumentation tests. # Android instrumentation test.
+ "additionalTestOptions": { # Optional. Additional test options to pass to the test runner. Passed to `am instrument` command as `-e` options, which will be passed to the instrumentation test runner using its `onCreate()` method. Formats supported in test_targets are not allowed to be used here. Limits: - Maximum number of entries: 32. - Maximum key size: 64 bytes (UTF-8). - Maximum value size: 1024 bytes (UTF-8).
+ "a_key": "A String",
+ },
+ "enableCodeCoverage": True or False, # Optional. Whether to enable code coverage collection for the test. A coverage file `coverage.ec` will be uploaded to the results folder. For this to work, your classes have to be instrumented offline (build time) by EMMA/JaCoCo.
+ "instrumentationTimeout": "A String", # Optional. The timeout of the instrumentation test. Default value: 5 min. Range: [1 min, 3 hours].
+ "orchestratorVersion": "A String", # Optional. The version of the Android Test Orchestrator to use for the test. The available orchestrator versions can be retrieved from the catalog service. If set to "auto", the default orchestrator is used. If not set, no orchestrator is used.
+ "smartSharding": { # The smart sharding strategy to split the job into multiple shards based on the test methods and their recorded execution time. # Optional. Smart sharding strategy to split the job into multiple shards based on the test methods and their execution time.
+ "maxShardCount": 42, # Optional. The maximum number of shards to create. If unset or less than 1, system-defined max limits are used. This limit takes precedence if the targeted_shard_duration cannot be satisfied. Limits: - For physical devices, the number of shards must be <= 20. - For virtual devices, the number of shards must be <= 200.
+ "targetedShardDuration": "A String", # Required. The targeted duration of each shard. Limits: - Must be at least 2 minutes. - Must be at most 3 hours. Shard duration is not guaranteed because smart sharding uses test case history and default durations which may not be accurate. Durations are calculated based on the following inputs: - Timing records from previous runs of the same test case. - For new test cases, the average duration of other known test cases. - A system-chosen, default duration if there are no previous timing records available. Because the actual shard duration can exceed the targeted shard duration, we recommend that you set the targeted value at least 5 minutes less than the maximum allowed instrumentation timeout. This approach avoids cancelling the shard before all tests can finish.
+ "timingRecord": { # Input file. # Required. The timing record file to use for smart sharding. If the file does not exist, smart sharding will use default test time (30s) for each test method to shard the job into multiple shards. This file will be overwritten with the latest timing record after the job is completed.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ },
+ "testInstallable": { # An Android Installable represents the file(s) for installing an Android package on a device. This can be an APK, an Android App Bundle (AAB), or an APK Set. # Required. The test package to install and run the test.
+ "files": [ # Required. Files that make up the package. Supported formats are distinguished by their file extension: - APK: One or more files with extension `.apk`. - App Bundle: A single file with extension `.aab`. - APK Set: A single file with extension `.apks`.
+ { # Input file.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ },
+ "testRunnerClass": "A String", # Optional. Full class name of the test runner class. The class must be `androidx.test.runner.AndroidJUnitRunner` or a subclass of it. The default value is determined by examining the application's manifest. If multiple instrumentations are found, the first one in the manifest will be used.
+ "testTargets": [ # Optional. A list of test targets or target filters to run. Each target must be fully qualified with the package name or class name, in one of these formats: - `package package_name` - `notPackage com.package.to.skip` - `class package_name.class_name` - `class package_name.class_name#method_name` - `notClass com.foo.ClassToSkip` - `notClass com.foo.ClassName#testMethodToSkip` - `annotation com.foo.AnnotationToRun` - `notAnnotation com.foo.AnnotationToSkip` - `size [small|medium|large]` Formats like `testfile` or `notTestfile` won't be supported. If empty, all targets in the module will be run. Limits: - Maximum number of entries: 1024.
+ "A String",
+ ],
+ "uniformSharding": { # Uniformly shards test cases given a total number of shards. It will be translated to `-e numShard` and `-e shardIndex` AndroidJUnitRunner arguments. With uniform sharding enabled, specifying either of these sharding arguments via `environment_variables` is invalid. Based on the sharding mechanism AndroidJUnitRunner uses, there is no guarantee that test cases will be distributed uniformly across all shards. # Optional. Uniform sharding strategy to split the job into multiple shards with equal number of test methods.
+ "shardCount": 42, # Required. The total number of shards to create. This must always be a positive number that is no greater than the total number of test cases. Limits: - For physical devices, the number of shards must be <= 20. - For virtual devices, the number of shards must be <= 200.
+ },
+ },
+ "androidNativeBinary": { # The configuration of an Android native binary execution. # Android native binary execution.
+ "androidNativeBinary": { # Input file. # Required. The file path of the Android native binary.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ "args": [ # Optional. Arguments for running the binary file. The flags will be appended to the command line that invokes the binary. Limits: - Maximum number of entries: 64 - Maximum entry size: 1024 bytes (UTF-8)
+ "A String",
+ ],
+ "envVars": { # Optional. A map of environment variables to set for the binary process. The keys are the variable names and the values are the variable values. Limits: - Maximum number of entries: 32 - Maximum key size: 64 bytes (UTF-8) - Key regex: `a-zA-Z_*` - Maximum value size: 1024 bytes (UTF-8)
+ "a_key": "A String",
+ },
+ "executionTimeout": "A String", # Optional. The timeout of the execution. Default value: 5 min. Range: [1 min, 3 hours].
+ },
+ "iosXcTest": { # The configuration of an iOS XCTest. # iOS XCTest.
+ "testsZip": { # Input file. # Required. The .zip containing the .xctestrun file and the contents of the DerivedData/Build/Products directory.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ "xcTestTimeout": "A String", # Optional. The timeout of the test. Default value: 5 min. Range: [1 min, 3 hours].
+ "xcodeVersion": "A String", # Optional. The Xcode version that should be used for the test. If not set, a system-default Xcode version is used. The available Xcode versions can be retrieved from the catalog service.
+ "xctestrun": { # Input file. # Optional. An .xctestrun file that will override the .xctestrun file in the tests zip.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ },
+ },
+ "allocationConfig": { # Allocation config. # Required. Allocation config.
+ "deviceConfigs": [ # Required. At least one device config is required. If more than one device config is required, the multiple devices are allocated to each shard of the OmniLab job to run multi-device-interaction tests.
+ { # The configuration of a run on a device.
+ "actions": [ # Optional. The actions to be performed on the device. Actions will be executed in the order they are specified in the list. Each action type can at most have 1 instance in the list.
+ { # The action to be performed on a device.
+ "androidBugreport": { # Captures a bugreport from the device. The output will be written to a file named `bugreport.zip` in the execution output directory. # Captures a bugreport from the device unless the test result is pass.
+ "collectOnPass": True or False, # Optional. Whether to deliver the bugreport when the test passes. If false, the bugreport is skipped on pass to save time (default behavior). If true, the bugreport is always delivered.
+ },
+ "androidDumpsys": { # Captures dumpsys output from the device. The output will be written to a file named `dumpsys.log` in the execution output directory. # Captures a dumpsys from the device.
+ "collectOnPass": True or False, # Optional. Whether to deliver the dumpsys when the test passes. If false, the dumpsys is skipped on pass to save time (default behavior). If true, the dumpsys is always delivered.
+ },
+ "androidInstallPackages": { # Installs Android packages on the device. At least one installable is specified when using this device action. Limits: - A maximum of 20 installables in total are allowed. - A maximum of 100 files are allowed in total across all installables. # Installs Android packages on the device.
+ "installables": [ # Optional. Deprecated: use `pre_target_app_installables`, `target_app` and `post_target_app_installables` instead. The Android packages to install on the device. The installation will be performed in the order specified, before the installables of all other fields.
+ { # An Android Installable represents the file(s) for installing an Android package on a device. This can be an APK, an Android App Bundle (AAB), or an APK Set.
+ "files": [ # Required. Files that make up the package. Supported formats are distinguished by their file extension: - APK: One or more files with extension `.apk`. - App Bundle: A single file with extension `.aab`. - APK Set: A single file with extension `.apks`.
+ { # Input file.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ },
+ ],
+ "postTargetAppInstallables": [ # Optional. The Android packages to install on the device after `target_app` (if specified) is installed. The installation will be performed in the order specified.
+ { # An Android Installable represents the file(s) for installing an Android package on a device. This can be an APK, an Android App Bundle (AAB), or an APK Set.
+ "files": [ # Required. Files that make up the package. Supported formats are distinguished by their file extension: - APK: One or more files with extension `.apk`. - App Bundle: A single file with extension `.aab`. - APK Set: A single file with extension `.apks`.
+ { # Input file.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ },
+ ],
+ "preTargetAppInstallables": [ # Optional. The Android packages to install on the device before `target_app` (if specified) is installed. The installation will be performed in the order specified.
+ { # An Android Installable represents the file(s) for installing an Android package on a device. This can be an APK, an Android App Bundle (AAB), or an APK Set.
+ "files": [ # Required. Files that make up the package. Supported formats are distinguished by their file extension: - APK: One or more files with extension `.apk`. - App Bundle: A single file with extension `.aab`. - APK Set: A single file with extension `.apks`.
+ { # Input file.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ },
+ ],
+ "targetApp": { # An Android Installable represents the file(s) for installing an Android package on a device. This can be an APK, an Android App Bundle (AAB), or an APK Set. # Optional. The primary Android package to install, serving as the target package for subsequent actions and as the installation ordering anchor. Whether this package is treated as the application under test depends on the job action: - Actions that require an explicit target package (such as performance metrics collection, or accessibility scans) use this package to identify the application to inspect or drive. - Actions that discover or manage targets independently (such as Android instrumentation tests, where target packages are defined in the test runner manifest) treat this field primarily as an installation order anchor between pre- and post-installables. Optional. If omitted, all packages in `pre_target_app_installables` and `post_target_app_installables` are installed without a designated target package.
+ "files": [ # Required. Files that make up the package. Supported formats are distinguished by their file extension: - APK: One or more files with extension `.apk`. - App Bundle: A single file with extension `.aab`. - APK Set: A single file with extension `.apks`.
+ { # Input file.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ },
+ },
+ "androidLogcat": { # Collects logcat output from the device. The output will be written to a file named `logcat.txt` in the execution output directory. # Collects logcat output from the device.
+ },
+ "androidMockLocation": { # Mocks the location of the Android device. # Mocks the location of the device.
+ "location": { # An object that represents a latitude/longitude pair. This is expressed as a pair of doubles to represent degrees latitude and degrees longitude. Unless specified otherwise, this object must conform to the WGS84 standard. Values must be within normalized ranges. # Required. The mock location to set on the device.
+ "latitude": 3.14, # The latitude in degrees. It must be in the range [-90.0, +90.0].
+ "longitude": 3.14, # The longitude in degrees. It must be in the range [-180.0, +180.0].
+ },
+ },
+ "androidOrientation": { # Sets the orientation of the device. # Sets the orientation of the device.
+ "orientation": "A String", # Required. The orientation to set the device to. One of `portrait` or `landscape`.
+ },
+ "androidPullFiles": { # Pulls directories and files from the device at the end of the run. Files will be copied to the '/artifacts' directory, with the absolute path structure preserved. Note that: 1. A clean device is provided for the run. 2. Any existing files in the output directory may be overwritten. 3. Pulling files is best effort. Will skip files if they don't exist on the device. # Pulls directories and files from the device at the end of the run.
+ "paths": [ # Required. Absolute directory or file paths to pull from the device. Limits: - A maximum of 10 paths are allowed.
+ "A String",
+ ],
+ },
+ "androidPushFiles": { # Pushes files to the device at the beginning of the run. Files are overwritten if a file with the same path already exists on the device, if device permissions allow. # Pushes files to the device at the beginning of the run.
+ "fileConfigs": [ # Required. Configs of pushing files to the device. Limits: - A maximum of 50 files are allowed.
+ { # The configuration of pushing a file to the device.
+ "destinationPath": "A String", # Required. The destination path on the device.
+ "sourceFile": { # Input file. # Required. The file to be pushed to the device.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ },
+ ],
+ },
+ "androidRecordVideo": { # Records a video of the device screen during the run. The video will be written to a file named `video.mp4` in the execution output directory. # Records a video of the device screen during the run.
+ "discardOnPass": True or False, # Optional. Whether to discard and not upload the recording when the test passes. Default is false.
+ },
+ "androidSwitchLocale": { # Switches the locale (language and region) of the device. # Switches the locale (language and region) of the device.
+ "localeCode": "A String", # Required. The locale (language and region) to switch the device to. The format is `language-region`, e.g. "en-US", "zh-CN", etc. The typical language value is a two or three-letter language code as defined in ISO639. The typical region value is a two-letter ISO 3166 code or a three-digit UN M.49 area code.
+ },
+ "iosAppPrivacyReport": { # Collects the exported iOS App Privacy Report during the test run. When enabled, the iOS device records application activity (such as network access, domain requests, and sensitive resource access like photos, camera, or location) and exports Apple's official App Privacy Report. The report will be written to a file named `AppActivityReport.ndjson` in the execution output directory. # Collects the exported iOS App Privacy Report during the run.
+ },
+ "iosInstallPackages": { # Installs iOS packages on the device. # Installs additional iOS packages on the device.
+ "ipas": [ # Required. Additional iOS packages (IPAs) to install on the device. Limits: - A maximum of 20 IPAs are allowed.
+ { # Input file.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ },
+ "iosPullFiles": { # Pulls directories and files from the iOS device sandbox at the end of the run. # Pulls directories and files from the iOS device sandbox at the end of the run.
+ "paths": [ # Required. Absolute directory or file paths to pull from the device. Limits: - A maximum of 10 paths are allowed.
+ { # The configuration of pulling a file or directory from the iOS device.
+ "bundleId": "A String", # Required. The bundle ID of the application sandbox.
+ "devicePath": "A String", # Required. The device path relative to the app sandbox, e.g. "/Documents/output/".
+ },
+ ],
+ },
+ "iosPushFiles": { # Pushes files to the iOS device sandbox at the beginning of the run. # Pushes files to the iOS device sandbox at the beginning of the run.
+ "fileConfigs": [ # Required. Configs of pushing files to the device. Limits: - A maximum of 50 files are allowed.
+ { # The configuration of pushing a file to the iOS device.
+ "bundleId": "A String", # Required. The bundle ID of the application sandbox.
+ "destinationPath": "A String", # Required. The destination path relative to the app sandbox, e.g. "/Documents/file.txt".
+ "sourceFile": { # Input file. # Required. The file to be pushed.
+ "gcsInputFile": { # A path to a file or directory in Google Cloud Storage. # An input file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ },
+ ],
+ },
+ "iosRecordVideo": { # Records a video of the iOS device screen during the run. The video will be written to a file named `video.mp4` in the execution output directory. # Records a video of the iOS device screen during the run.
+ "discardOnPass": True or False, # Optional. Whether to discard the video if the test passes. If not specified, the default is false (always keep the video).
+ },
+ "iosSwitchLocale": { # Switches the locale (language and region) of the iOS application. # Switches the locale (language and region) of the iOS application.
+ "localeCode": "A String", # Required. The locale (language and region) to switch the app to. The format is `language-region` or `language`, e.g. "en-US", "zh-CN", "ja", etc.
+ },
+ },
+ ],
+ "requirement": { # The requirement of a device. # Required. The requirement of the device.
+ "deviceId": "A String", # The device ID of a device in the catalog. The device ID is the last part of a device's resource name.
+ },
+ },
+ ],
+ },
+ "displayName": "A String", # Optional. User-settable, human-readable name for the job. If set, it must be unique within the session. If not set, the display name will default to `job-`, where `` is the 0-based index of the job in the session formatted as three digits (e.g., job-000, job-001, ...). Maximum size is 63 bytes when encoded as UTF-8. If set, must match regex: `^A-Za-z0-9*$`.
+ "labels": { # Optional. User-defined metadata for tracking or categorization. These labels do not affect job execution and are surfaced in the JobReport. Limits: - Maximum number of entries: 16. - Maximum key size: 32 bytes (UTF-8). - Maximum value size: 1024 bytes (UTF-8).
+ "a_key": "A String",
+ },
+ "settings": { # Job settings to control the job execution. # Optional. Job settings.
+ "retrySettings": { # Retry settings. # Optional. The retry settings of the job.
+ "flakyTestRetryStrategy": { # Default retry strategy. It will retry on test failures for up to flaky_test_attempts (including the initial run). It also retries on infra issues for up to 2 attempts (including the initial run). So in total, an execution can run up to flaky_test_attempts * 2 times in the worst case. # Optional. The default retry strategy. Allows an Execution to retry on test failures and infrastructure errors.
+ "flakyTestAttempts": 42, # Required. The total attempts for flaky tests, including the initial run. Default value: 1 (no retry). Range: [1, 5].
+ "parallelRetry": True or False, # Optional. Whether to retry the test failures in parallel. By default, the test is retried sequentially. If true, when the initial attempt fails, (flaky_test_attempts - 1) attempts will be triggered at the same time to run in parallel.
+ "testReductionMode": "A String", # Optional. The mode of test reduction for retry. If the test runner doesn't support the specified test reduction mode, the request will be rejected with an `INVALID_ARGUMENT` error.
+ },
+ },
+ },
+ },
+ ],
+ "notificationConfig": { # Config to control session notification. # Optional. Notification config for the session.
+ "pubsubTopic": [ # Optional. The Pub/Sub topics to which session events are published. Format: `projects/{project}/topics/{topic}`. See https://cloud.google.com/pubsub/docs/admin#topic_and_subscription_name_restrictions
+ "A String",
+ ],
+ },
+ "outputDirectoryConfig": { # Config to control session output file directory. # Required. Output file directory config for the session.
+ "flatDirectoryStructure": True or False, # Optional. Whether to write output files directly under the output directory instead of nesting them under service-generated subdirectories. By default (`false`), output files are stored under `////`. When `true`, the session ID subdirectory is never appended, and the job display name subdirectory is appended only when the session has more than one job. Output files are therefore stored under: - `//` for a single-job session. - `///` for a multi-job session. Set this to `true` when the output directory is already unique per session (for example, when a CI system generates it), to avoid redundant nesting.
+ "gcsOutputDirectory": { # A path to a file or directory in Google Cloud Storage. # The Google Cloud Storage path of the output directory (e.g. `gs://my-bucket/output`). The bucket must exist. If the bucket is located in another project or uses fine-grained access controls, ensure the Device Run Service Agent of the project (`service-@gcp-sa-devicerun.iam.gserviceaccount.com`) is granted access to the bucket (such as `roles/storage.objectUser`).
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ },
+ "sessionReport": { # The runtime information and result report of a session. # Output only. The runtime information and result report of the session.
+ "endTime": "A String", # Output only. The end time of the session.
+ "id": "A String", # Output only. The unique identifier of the session.
+ "jobReports": [ # Output only. Reports of the jobs in the session.
+ { # The runtime information and result report of a job.
+ "displayName": "A String", # Output only. The display_name set by users in the JobConfig.
+ "endTime": "A String", # Output only. The end time of the job.
+ "executionReports": [ # Output only. Reports of the execution attempts of the job.
+ { # The runtime information and result report of a single on-device execution attempt.
+ "displayName": "A String", # Output only. The display_name set by users in the ExecutionConfig.
+ "endTime": "A String", # Output only. The end time of the execution.
+ "id": "A String", # Output only. The unique identifier of the execution.
+ "outputFiles": [ # Output only. The output files of the execution.
+ { # Output file.
+ "gcsOutputFile": { # A path to a file or directory in Google Cloud Storage. # An output file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ "result": { # The result of a session/job/execution. # Output only. The result of the execution.
+ "cause": { # Describes the cause of the non-passed result occurred during the execution. # Output only. Detailed result cause diagnostics. Set if type is not PASSED.
+ "summary": { # Describes the summary of an issue (error or warning) with structured details. # Output only. Structured cause detail.
+ "message": "A String", # Output only. Human-readable explanation of the issue in English.
+ "reason": "A String", # Output only. The reason of the issue. This is a constant value that identifies the proximate cause of the issue. This should be at most 63 characters and match a regular expression of `A-Z*[A-Z0-9]`, which represents UPPER_SNAKE_CASE.
+ "type": "A String", # Output only. The issue classification based on responsibility.
+ },
+ },
+ "resultType": "A String", # Output only. The result type of the session/job/execution.
+ },
+ "startTime": "A String", # Output only. The start time of the execution.
+ "status": { # The status of a session/job/execution. # Output only. The status of the execution.
+ "progressMessages": [ # Output only. Human-readable, detailed descriptions of the session/job/execution's progress. For example: "Provisioning a device", "Starting Test". Each message should contain only one line of text. During the course of execution new data may be appended to the end of progress_messages.
+ "A String",
+ ],
+ "statusType": "A String", # Output only. The status type of the session/job/execution.
+ },
+ "warnings": [ # Output only. Non-fatal warnings collected during the execution.
+ { # Non-fatal operational anomaly, lint observation, or execution insight.
+ "summary": { # Describes the summary of an issue (error or warning) with structured details. # Output only. Detailed warning summary.
+ "message": "A String", # Output only. Human-readable explanation of the issue in English.
+ "reason": "A String", # Output only. The reason of the issue. This is a constant value that identifies the proximate cause of the issue. This should be at most 63 characters and match a regular expression of `A-Z*[A-Z0-9]`, which represents UPPER_SNAKE_CASE.
+ "type": "A String", # Output only. The issue classification based on responsibility.
+ },
+ },
+ ],
+ },
+ ],
+ "id": "A String", # Output only. The unique identifier of the job.
+ "labels": { # Output only. The original labels provided by the user during job creation.
+ "a_key": "A String",
+ },
+ "outputFiles": [ # Output only. The output files of the job.
+ { # Output file.
+ "gcsOutputFile": { # A path to a file or directory in Google Cloud Storage. # An output file in Google Cloud Storage.
+ "path": "A String", # Required. The Google Cloud Storage path of the file or directory. Format: `gs:///`.
+ },
+ },
+ ],
+ "result": { # The result of a session/job/execution. # Output only. The result of the job.
+ "cause": { # Describes the cause of the non-passed result occurred during the execution. # Output only. Detailed result cause diagnostics. Set if type is not PASSED.
+ "summary": { # Describes the summary of an issue (error or warning) with structured details. # Output only. Structured cause detail.
+ "message": "A String", # Output only. Human-readable explanation of the issue in English.
+ "reason": "A String", # Output only. The reason of the issue. This is a constant value that identifies the proximate cause of the issue. This should be at most 63 characters and match a regular expression of `A-Z*[A-Z0-9]`, which represents UPPER_SNAKE_CASE.
+ "type": "A String", # Output only. The issue classification based on responsibility.
+ },
+ },
+ "resultType": "A String", # Output only. The result type of the session/job/execution.
+ },
+ "startTime": "A String", # Output only. The start time of the job.
+ "status": { # The status of a session/job/execution. # Output only. The status of the job.
+ "progressMessages": [ # Output only. Human-readable, detailed descriptions of the session/job/execution's progress. For example: "Provisioning a device", "Starting Test". Each message should contain only one line of text. During the course of execution new data may be appended to the end of progress_messages.
+ "A String",
+ ],
+ "statusType": "A String", # Output only. The status type of the session/job/execution.
+ },
+ "warnings": [ # Output only. Non-fatal warnings collected during the job.
+ { # Non-fatal operational anomaly, lint observation, or execution insight.
+ "summary": { # Describes the summary of an issue (error or warning) with structured details. # Output only. Detailed warning summary.
+ "message": "A String", # Output only. Human-readable explanation of the issue in English.
+ "reason": "A String", # Output only. The reason of the issue. This is a constant value that identifies the proximate cause of the issue. This should be at most 63 characters and match a regular expression of `A-Z*[A-Z0-9]`, which represents UPPER_SNAKE_CASE.
+ "type": "A String", # Output only. The issue classification based on responsibility.
+ },
+ },
+ ],
+ },
+ ],
+ "result": { # The result of a session/job/execution. # Output only. The result of the session.
+ "cause": { # Describes the cause of the non-passed result occurred during the execution. # Output only. Detailed result cause diagnostics. Set if type is not PASSED.
+ "summary": { # Describes the summary of an issue (error or warning) with structured details. # Output only. Structured cause detail.
+ "message": "A String", # Output only. Human-readable explanation of the issue in English.
+ "reason": "A String", # Output only. The reason of the issue. This is a constant value that identifies the proximate cause of the issue. This should be at most 63 characters and match a regular expression of `A-Z*[A-Z0-9]`, which represents UPPER_SNAKE_CASE.
+ "type": "A String", # Output only. The issue classification based on responsibility.
+ },
+ },
+ "resultType": "A String", # Output only. The result type of the session/job/execution.
+ },
+ "startTime": "A String", # Output only. The start time of the session.
+ "status": { # The status of a session/job/execution. # Output only. The status of the session.
+ "progressMessages": [ # Output only. Human-readable, detailed descriptions of the session/job/execution's progress. For example: "Provisioning a device", "Starting Test". Each message should contain only one line of text. During the course of execution new data may be appended to the end of progress_messages.
+ "A String",
+ ],
+ "statusType": "A String", # Output only. The status type of the session/job/execution.
+ },
+ },
+ },
+ ],
+}
+cancel(name, body=None, x__xgafv=None)
Cancels an in-progress automation session. This RPC returns immediately and cancellation proceeds asynchronously. If the session is already finished, this RPC will have no effect. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.servingConfigs.html b/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.servingConfigs.html index fb22b0f9a9..d33db79dc1 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.servingConfigs.html @@ -1454,6 +1454,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.userEvents.html b/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.userEvents.html index 1741f3eda6..e25b9cf203 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.userEvents.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.userEvents.html @@ -204,6 +204,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -431,6 +432,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -583,6 +585,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.widgetConfigs.html b/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.widgetConfigs.html index 11ea95e992..ad243504c1 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.widgetConfigs.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.collections.dataStores.widgetConfigs.html @@ -354,6 +354,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -643,6 +644,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -931,6 +933,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.servingConfigs.html b/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.servingConfigs.html index f09591cdaa..356ed95ee4 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.servingConfigs.html @@ -1454,6 +1454,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.widgetConfigs.html b/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.widgetConfigs.html index 7948576dd7..92fd7411dd 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.widgetConfigs.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.collections.engines.widgetConfigs.html @@ -354,6 +354,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -643,6 +644,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -931,6 +933,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.dataStores.servingConfigs.html b/docs/dyn/discoveryengine_v1.projects.locations.dataStores.servingConfigs.html index 12aac679b8..1407eb2f6e 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.dataStores.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.dataStores.servingConfigs.html @@ -1454,6 +1454,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.dataStores.userEvents.html b/docs/dyn/discoveryengine_v1.projects.locations.dataStores.userEvents.html index 6e1b91943c..2f2aed06f4 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.dataStores.userEvents.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.dataStores.userEvents.html @@ -204,6 +204,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -431,6 +432,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -583,6 +585,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.dataStores.widgetConfigs.html b/docs/dyn/discoveryengine_v1.projects.locations.dataStores.widgetConfigs.html index 0b3a30b8cb..60114178d0 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.dataStores.widgetConfigs.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.dataStores.widgetConfigs.html @@ -354,6 +354,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -643,6 +644,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -931,6 +933,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. diff --git a/docs/dyn/discoveryengine_v1.projects.locations.userEvents.html b/docs/dyn/discoveryengine_v1.projects.locations.userEvents.html index 5a30217571..39ddb5c1e3 100644 --- a/docs/dyn/discoveryengine_v1.projects.locations.userEvents.html +++ b/docs/dyn/discoveryengine_v1.projects.locations.userEvents.html @@ -201,6 +201,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -385,6 +386,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -537,6 +539,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.servingConfigs.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.servingConfigs.html index 1682049ac4..ec00527f70 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.servingConfigs.html @@ -1714,6 +1714,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.userEvents.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.userEvents.html index 678acf41b5..7f4f9b425f 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.userEvents.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.userEvents.html @@ -241,6 +241,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -505,6 +506,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -694,6 +696,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.widgetConfigs.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.widgetConfigs.html index b9f620ab18..56a7d504e1 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.widgetConfigs.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.dataStores.widgetConfigs.html @@ -359,6 +359,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -653,6 +654,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -946,6 +948,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.html index b7a136cb07..87e7cfd28b 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.html @@ -84,6 +84,11 @@Instance Methods
Returns the operations Resource.
++
+toolCalls()+Returns the toolCalls Resource.
+Close httplib2 connections.
diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.toolCalls.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.toolCalls.html new file mode 100644 index 0000000000..f3c39b2b05 --- /dev/null +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.toolCalls.html @@ -0,0 +1,91 @@ + + + +Discovery Engine API . projects . locations . collections . engines . assistants . agents . toolCalls
+Instance Methods
++
+operations()+Returns the operations Resource.
+ ++
+close()Close httplib2 connections.
+Method Details
+++ + \ No newline at end of file diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.toolCalls.operations.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.toolCalls.operations.html new file mode 100644 index 0000000000..3a1ea5dda3 --- /dev/null +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.assistants.agents.toolCalls.operations.html @@ -0,0 +1,124 @@ + + + +close()+Close httplib2 connections.+Discovery Engine API . projects . locations . collections . engines . assistants . agents . toolCalls . operations
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service.
+Method Details
+++ +close()+Close httplib2 connections.+++ + \ No newline at end of file diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.servingConfigs.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.servingConfigs.html index 99027036cd..2651c81505 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.servingConfigs.html @@ -1714,6 +1714,7 @@get(name, x__xgafv=None)+Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service. + +Args: + name: string, The name of the operation resource. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.widgetConfigs.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.widgetConfigs.html index e31370f965..94b301326a 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.widgetConfigs.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.collections.engines.widgetConfigs.html @@ -359,6 +359,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -653,6 +654,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -946,6 +948,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.servingConfigs.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.servingConfigs.html index 94dc72e55f..c6745db009 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.servingConfigs.html @@ -1714,6 +1714,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.userEvents.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.userEvents.html index 385e610325..c4ce56645d 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.userEvents.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.userEvents.html @@ -241,6 +241,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -505,6 +506,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -694,6 +696,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.widgetConfigs.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.widgetConfigs.html index 30ed0180cc..f10fb4fe36 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.widgetConfigs.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.dataStores.widgetConfigs.html @@ -359,6 +359,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -653,6 +654,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. @@ -946,6 +948,7 @@Method Details
"googleDrivePickerEnabled": True or False, # Output only. Whether the Google Drive file picker is available to end-users. Declared `optional` for the same field-presence reason as `onedrive_picker_enabled` above. "interactionType": "A String", # Describes widget (or web app) interaction type "modelConfigInfo": { # The resolved, server-side view of model selector configuration for the end-user. The backend computes this per-request by applying, in order: Mendel flag evaluation, regional availability rules based on the engine's location, and admin-panel overrides from `model_configs`. The backend is the single source of truth for this configuration; clients should render `resolved_models` directly in the model selector dropdown, in the order provided, without applying their own filtering, ordering, or localization. # Output only. The resolved, server-side view of model selector configuration. Holds both the ordered list of models that should appear in the model selector dropdown and the model that should be selected by default. Clients should render this directly without applying their own filtering, ordering, or localization. The legacy `model_configs` map above is retained for backward compatibility with clients that have not yet migrated to consuming this field. + "autoModelId": "A String", # Output only. The `model_id` a client must send when the end-user picks the "Auto" entry (the empty `model_id`). Populated only for surfaces known to have no true "Auto" mode, where leaving the model unset would fall back to a platform default rather than a backend-chosen model. Empty on every other surface, where "Auto" is resolved server-side, and when none of the surface's ranked models is available. When set, it is always one of the non-empty `model_id`s present in `resolved_models`. "defaultModelId": "A String", # Output only. The `model_id` of the model that should be selected by default in the model selector when the end-user has not made an explicit choice. The value is always one of the `model_id`s present in `resolved_models`. "resolvedModels": [ # Output only. The list of models that are available to the end-user in the model selector, in the order in which they should be displayed. { # A single model that is resolved to be available to the end-user in the model selector, with all of its localized display metadata. diff --git a/docs/dyn/discoveryengine_v1alpha.projects.locations.userEvents.html b/docs/dyn/discoveryengine_v1alpha.projects.locations.userEvents.html index a59ea3593f..7174a0a254 100644 --- a/docs/dyn/discoveryengine_v1alpha.projects.locations.userEvents.html +++ b/docs/dyn/discoveryengine_v1alpha.projects.locations.userEvents.html @@ -238,6 +238,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -459,6 +460,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -648,6 +650,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.servingConfigs.html b/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.servingConfigs.html index b2cbd62e31..679380e9c2 100644 --- a/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.servingConfigs.html @@ -1551,6 +1551,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.userEvents.html b/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.userEvents.html index 0e2a8ad70a..30752db899 100644 --- a/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.userEvents.html +++ b/docs/dyn/discoveryengine_v1beta.projects.locations.collections.dataStores.userEvents.html @@ -204,6 +204,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -431,6 +432,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -583,6 +585,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1beta.projects.locations.collections.engines.servingConfigs.html b/docs/dyn/discoveryengine_v1beta.projects.locations.collections.engines.servingConfigs.html index e996f8b2ff..93339632c8 100644 --- a/docs/dyn/discoveryengine_v1beta.projects.locations.collections.engines.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1beta.projects.locations.collections.engines.servingConfigs.html @@ -1551,6 +1551,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.servingConfigs.html b/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.servingConfigs.html index 769b1b38e3..ef12908aaf 100644 --- a/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.servingConfigs.html +++ b/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.servingConfigs.html @@ -1551,6 +1551,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.userEvents.html b/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.userEvents.html index fc8db23504..2ffe557538 100644 --- a/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.userEvents.html +++ b/docs/dyn/discoveryengine_v1beta.projects.locations.dataStores.userEvents.html @@ -204,6 +204,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -431,6 +432,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -583,6 +585,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/discoveryengine_v1beta.projects.locations.userEvents.html b/docs/dyn/discoveryengine_v1beta.projects.locations.userEvents.html index 9834b60b14..b858947e56 100644 --- a/docs/dyn/discoveryengine_v1beta.projects.locations.userEvents.html +++ b/docs/dyn/discoveryengine_v1beta.projects.locations.userEvents.html @@ -201,6 +201,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -385,6 +386,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. @@ -537,6 +539,7 @@Method Details
}, "questionIndex": 42, # The index of the user input within the conversation messages. "session": "A String", # Name of the newly generated or continued session. + "taskName": "A String", # Optional. The full resource name of the Sobi task if the conversation was handled by a long-running agent task. Format: projects/{project}/locations/{location}/tasks/{task_id} }, "dataTermsAccepted": True or False, # Optional. Whether the customer accepted data use terms. "feedbackSource": "A String", # Optional. The UI component the user feedback comes from, which could be GOOGLE_CONSOLE, GOOGLE_WIDGET, GOOGLE_WEBAPP. diff --git a/docs/dyn/displayvideo_v2.advertisers.html b/docs/dyn/displayvideo_v2.advertisers.html index 66ecfe5ab1..d4f9cb96df 100644 --- a/docs/dyn/displayvideo_v2.advertisers.html +++ b/docs/dyn/displayvideo_v2.advertisers.html @@ -485,7 +485,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -506,7 +506,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -565,7 +565,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -798,7 +798,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -819,7 +819,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -878,7 +878,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1266,7 +1266,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1287,7 +1287,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1346,7 +1346,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.advertisers.insertionOrders.html b/docs/dyn/displayvideo_v2.advertisers.insertionOrders.html index c1ad894f77..42ea287b96 100644 --- a/docs/dyn/displayvideo_v2.advertisers.insertionOrders.html +++ b/docs/dyn/displayvideo_v2.advertisers.insertionOrders.html @@ -134,6 +134,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "billableOutcome": "A String", # Immutable. The billable outcome of the insertion order. Outcome based buying is deprecated. `BILLABLE_OUTCOME_PAY_PER_IMPRESSION` is the only valid value. "budget": { # Settings that control how insertion order budget is allocated. # Required. The budget allocation settings of the insertion order. @@ -233,6 +239,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "billableOutcome": "A String", # Immutable. The billable outcome of the insertion order. Outcome based buying is deprecated. `BILLABLE_OUTCOME_PAY_PER_IMPRESSION` is the only valid value. "budget": { # Settings that control how insertion order budget is allocated. # Required. The budget allocation settings of the insertion order. @@ -359,6 +371,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "billableOutcome": "A String", # Immutable. The billable outcome of the insertion order. Outcome based buying is deprecated. `BILLABLE_OUTCOME_PAY_PER_IMPRESSION` is the only valid value. "budget": { # Settings that control how insertion order budget is allocated. # Required. The budget allocation settings of the insertion order. @@ -471,6 +489,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "billableOutcome": "A String", # Immutable. The billable outcome of the insertion order. Outcome based buying is deprecated. `BILLABLE_OUTCOME_PAY_PER_IMPRESSION` is the only valid value. "budget": { # Settings that control how insertion order budget is allocated. # Required. The budget allocation settings of the insertion order. @@ -590,6 +614,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "billableOutcome": "A String", # Immutable. The billable outcome of the insertion order. Outcome based buying is deprecated. `BILLABLE_OUTCOME_PAY_PER_IMPRESSION` is the only valid value. "budget": { # Settings that control how insertion order budget is allocated. # Required. The budget allocation settings of the insertion order. @@ -690,6 +720,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "billableOutcome": "A String", # Immutable. The billable outcome of the insertion order. Outcome based buying is deprecated. `BILLABLE_OUTCOME_PAY_PER_IMPRESSION` is the only valid value. "budget": { # Settings that control how insertion order budget is allocated. # Required. The budget allocation settings of the insertion order. diff --git a/docs/dyn/displayvideo_v2.advertisers.lineItems.html b/docs/dyn/displayvideo_v2.advertisers.lineItems.html index 99b445aff7..28f3893628 100644 --- a/docs/dyn/displayvideo_v2.advertisers.lineItems.html +++ b/docs/dyn/displayvideo_v2.advertisers.lineItems.html @@ -245,7 +245,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -266,7 +266,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -325,7 +325,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -602,7 +602,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -623,7 +623,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -682,7 +682,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -833,6 +833,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -954,7 +960,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -976,8 +982,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` @@ -1104,6 +1110,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -1225,7 +1237,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -1247,8 +1259,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` @@ -1335,6 +1347,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -1456,7 +1474,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -1478,8 +1496,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` @@ -1621,6 +1639,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -1742,7 +1766,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -1764,8 +1788,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` @@ -1865,6 +1889,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -1986,7 +2016,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -2008,8 +2038,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` @@ -2116,6 +2146,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -2237,7 +2273,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -2259,8 +2295,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` @@ -2348,6 +2384,12 @@Method Details
"performanceGoalAmountMicros": "A String", # Required. The performance goal the bidding strategy will attempt to meet or beat, in micros of the advertiser's currency or in micro of the ROAS (Return On Advertising Spend) value which is also based on advertiser's currency. Must be greater than or equal to a billable unit of the given currency and smaller or equal to upper bounds. Each performance_goal_type has its upper bound: * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA`, upper bound is 10000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, upper bound is 1000.00 USD. * when performance_goal_type is `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`, upper bound is 1000.00 and lower bound is 0.01. Example: If set to `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM`, the bid price will be based on the probability that each available impression will be viewable. For example, if viewable CPM target is $2 and an impression is 40% likely to be viewable, the bid price will be $0.80 CPM (40% of $2). For example, 1500000 represents 1.5 standard units of the currency or ROAS value. "performanceGoalType": "A String", # Required. The type of the performance goal that the bidding strategy will try to meet or beat. For line item level usage, the value must be one of: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CPC` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CUSTOM_ALGO`. }, + "youtubeAndPartnersBid": { # Settings that control the bid strategy for YouTube and Partners resources. # A bid strategy used by YouTube and Partners resources. It can only be used for a YouTube and Partners line item or ad group entity. + "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. + "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. + "type": "A String", # The type of the bidding strategy. + "value": "A String", # The value used by the bidding strategy. When the bidding strategy is assigned at the line item level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` When the bidding strategy is assigned at the ad group level, this field is only applicable for the following strategy types: * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_MANUAL_CPV` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPA` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_RESERVE_CPM` * `YOUTUBE_AND_PARTNERS_BIDDING_STRATEGY_TYPE_TARGET_ROAS` If not using an applicable strategy, the value of this field will be 0. + }, }, "budget": { # Settings that control how budget is allocated. # Required. The budget allocation setting of the line item. "budgetAllocationType": "A String", # Required. The type of the budget allocation. `LINE_ITEM_BUDGET_ALLOCATION_TYPE_AUTOMATIC` is only applicable when automatic budget allocation is enabled for the parent insertion order. This field must be set to `LINE_ITEM_BUDGET_ALLOCATION_TYPE_FIXED` for Demand Gen line items. @@ -2469,7 +2511,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "biddingStrategy": { # Settings that control the bid strategy for YouTube and Partners resources. # Required. The bidding strategy of the YouTube and Partners line item. "adGroupEffectiveTargetCpaSource": "A String", # Output only. Source of the effective target CPA value for ad group. "adGroupEffectiveTargetCpaValue": "A String", # Output only. The effective target CPA for ad group, in micros of advertiser's currency. @@ -2491,8 +2533,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementSettings": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement settings of the line item. "brandLiftVendorConfigs": [ # The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` diff --git a/docs/dyn/displayvideo_v2.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v2.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html index 01db811888..83b4d36a35 100644 --- a/docs/dyn/displayvideo_v2.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v2.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html @@ -138,7 +138,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -276,7 +275,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -297,7 +296,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -356,7 +355,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -574,7 +573,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -595,7 +594,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -654,7 +653,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -789,7 +788,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -861,7 +859,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1005,7 +1002,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1026,7 +1023,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1085,7 +1082,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1220,7 +1217,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1369,7 +1365,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1390,7 +1386,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1449,7 +1445,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.advertisers.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v2.advertisers.targetingTypes.assignedTargetingOptions.html index e9eebd1ca7..7674b681f6 100644 --- a/docs/dyn/displayvideo_v2.advertisers.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v2.advertisers.targetingTypes.assignedTargetingOptions.html @@ -137,7 +137,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -275,7 +274,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -296,7 +295,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -355,7 +354,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -573,7 +572,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -594,7 +593,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -653,7 +652,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -787,7 +786,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -858,7 +856,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1002,7 +999,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1023,7 +1020,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1082,7 +1079,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1216,7 +1213,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1365,7 +1361,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1386,7 +1382,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1445,7 +1441,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroupAds.html b/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroupAds.html index e6f58c14ae..d1528f6935 100644 --- a/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroupAds.html +++ b/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroupAds.html @@ -143,7 +143,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -171,7 +171,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -221,7 +221,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -342,7 +342,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -370,7 +370,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -420,7 +420,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. diff --git a/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.html b/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.html index 65539a74ee..76ccd2a3c1 100644 --- a/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.html +++ b/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.html @@ -237,7 +237,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -258,7 +258,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -317,7 +317,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.targetingTypes.assignedTargetingOptions.html index 16865ce4f7..383468d329 100644 --- a/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v2.advertisers.youtubeAdGroups.targetingTypes.assignedTargetingOptions.html @@ -132,7 +132,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -276,7 +275,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -297,7 +296,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -356,7 +355,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -491,7 +490,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -640,7 +638,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -661,7 +659,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -720,7 +718,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.partners.html b/docs/dyn/displayvideo_v2.partners.html index 48b7640da2..4cae5cecd1 100644 --- a/docs/dyn/displayvideo_v2.partners.html +++ b/docs/dyn/displayvideo_v2.partners.html @@ -234,7 +234,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -255,7 +255,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -314,7 +314,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -547,7 +547,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -568,7 +568,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -627,7 +627,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.partners.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v2.partners.targetingTypes.assignedTargetingOptions.html index ab26676d6d..6e6a199e0a 100644 --- a/docs/dyn/displayvideo_v2.partners.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v2.partners.targetingTypes.assignedTargetingOptions.html @@ -137,7 +137,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -275,7 +274,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -296,7 +295,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -355,7 +354,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -573,7 +572,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -594,7 +593,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -653,7 +652,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -787,7 +786,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -858,7 +856,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1002,7 +999,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1023,7 +1020,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1082,7 +1079,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1216,7 +1213,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1365,7 +1361,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1386,7 +1382,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1445,7 +1441,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v2.targetingTypes.targetingOptions.html b/docs/dyn/displayvideo_v2.targetingTypes.targetingOptions.html index a88a13171d..9203dc557d 100644 --- a/docs/dyn/displayvideo_v2.targetingTypes.targetingOptions.html +++ b/docs/dyn/displayvideo_v2.targetingTypes.targetingOptions.html @@ -136,7 +136,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -258,7 +257,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. @@ -315,7 +314,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -443,7 +441,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. @@ -516,7 +514,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -658,7 +655,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. diff --git a/docs/dyn/displayvideo_v3.advertisers.adGroupAds.html b/docs/dyn/displayvideo_v3.advertisers.adGroupAds.html index 3e0ca785c3..67947b9c22 100644 --- a/docs/dyn/displayvideo_v3.advertisers.adGroupAds.html +++ b/docs/dyn/displayvideo_v3.advertisers.adGroupAds.html @@ -279,7 +279,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -502,7 +502,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -552,7 +552,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -792,7 +792,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1015,7 +1015,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1065,7 +1065,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1332,7 +1332,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1555,7 +1555,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1605,7 +1605,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1858,7 +1858,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2081,7 +2081,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2131,7 +2131,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2391,7 +2391,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2614,7 +2614,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2664,7 +2664,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2905,7 +2905,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -3128,7 +3128,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -3178,7 +3178,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. diff --git a/docs/dyn/displayvideo_v3.advertisers.adGroups.html b/docs/dyn/displayvideo_v3.advertisers.adGroups.html index 6a94f42fd3..43d6b86c8d 100644 --- a/docs/dyn/displayvideo_v3.advertisers.adGroups.html +++ b/docs/dyn/displayvideo_v3.advertisers.adGroups.html @@ -265,7 +265,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -286,7 +286,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -345,7 +345,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -663,7 +663,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -684,7 +684,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -743,7 +743,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -952,8 +952,6 @@Method Details
"productMatchType": "A String", # How products are selected by the product feed. }, "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # Optional. The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the ad group. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1029,8 +1027,6 @@Method Details
"productMatchType": "A String", # How products are selected by the product feed. }, "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # Optional. The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the ad group. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1133,8 +1129,6 @@Method Details
"productMatchType": "A String", # How products are selected by the product feed. }, "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # Optional. The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the ad group. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1223,8 +1217,6 @@Method Details
"productMatchType": "A String", # How products are selected by the product feed. }, "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # Optional. The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the ad group. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1320,8 +1312,6 @@Method Details
"productMatchType": "A String", # How products are selected by the product feed. }, "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # Optional. The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the ad group. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1398,8 +1388,6 @@Method Details
"productMatchType": "A String", # How products are selected by the product feed. }, "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # Optional. The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the ad group. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, diff --git a/docs/dyn/displayvideo_v3.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v3.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html index bee3a328f7..844e289c24 100644 --- a/docs/dyn/displayvideo_v3.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v3.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html @@ -138,7 +138,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -300,7 +299,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -321,7 +320,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -380,7 +379,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -641,7 +640,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -662,7 +661,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -721,7 +720,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -876,7 +875,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -949,7 +947,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1117,7 +1114,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1138,7 +1135,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1197,7 +1194,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1352,7 +1349,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1525,7 +1521,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1546,7 +1542,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1605,7 +1601,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v3.advertisers.html b/docs/dyn/displayvideo_v3.advertisers.html index 0b11a370da..9839bf0ef6 100644 --- a/docs/dyn/displayvideo_v3.advertisers.html +++ b/docs/dyn/displayvideo_v3.advertisers.html @@ -509,7 +509,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -530,7 +530,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -589,7 +589,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -865,7 +865,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -886,7 +886,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -945,7 +945,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1382,7 +1382,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1403,7 +1403,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1462,7 +1462,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v3.advertisers.lineItems.html b/docs/dyn/displayvideo_v3.advertisers.lineItems.html index 0d2b6ecdac..c17e08a1fd 100644 --- a/docs/dyn/displayvideo_v3.advertisers.lineItems.html +++ b/docs/dyn/displayvideo_v3.advertisers.lineItems.html @@ -268,7 +268,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -289,7 +289,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -348,7 +348,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -668,7 +668,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -689,7 +689,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -748,7 +748,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1038,8 +1038,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1047,7 +1045,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1062,8 +1060,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -1309,8 +1307,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1318,7 +1314,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1333,8 +1329,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -1540,8 +1536,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1549,7 +1543,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1564,8 +1558,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -1826,8 +1820,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -1835,7 +1827,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1850,8 +1842,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -2070,8 +2062,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -2079,7 +2069,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -2094,8 +2084,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -2321,8 +2311,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -2330,7 +2318,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -2345,8 +2333,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -2553,8 +2541,6 @@Method Details
}, "reservationType": "A String", # Output only. The reservation type of the line item. "targetingExpansion": { # Settings that control the [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. # The [optimized targeting](//support.google.com/displayvideo/answer/12060859) settings of the line item. This config is only applicable for display, video, or audio line items that use automated bidding and positively target eligible audience lists. - "audienceExpansionLevel": "A String", # Output only. Magnitude of expansion for eligible first-party user lists under this ad group. This field only applies to YouTube and Partners line item and ad group resources. - "audienceExpansionSeedListExcluded": True or False, # Output only. Whether to exclude seed list for audience expansion. This field only applies to YouTube and Partners line item and ad group resources. "enableOptimizedTargeting": True or False, # Required. Whether to enable Optimized Targeting for the line item. Optimized targeting is not compatible with all bid strategies. Attempting to set this field to `true` for a line item using the BiddingStrategy field fixed_bid or one of the following combinations of BiddingStrategy fields and BiddingStrategyPerformanceGoalType will result in an error: maximize_auto_spend_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_CIVA` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_IVO_TEN` * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_AV_VIEWED` performance_goal_auto_bid: * `BIDDING_STRATEGY_PERFORMANCE_GOAL_TYPE_VIEWABLE_CPM` This also applies if the line item inherits one of the above bid strategies from the parent insertion order. Bid strategies set at the insertion order-level will be inherited by their line items if the `InsertionOrder` budget field automationType is set to `INSERTION_ORDER_AUTOMATION_TYPE_BUDGET` or `INSERTION_ORDER_AUTOMATION_TYPE_BID_BUDGET`. "excludeDemographicExpansion": True or False, # Optional. Whether to exclude demographic expansion for Optimized Targeting. This field can only be set for Demand Gen ad groups. }, @@ -2562,7 +2548,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -2577,8 +2563,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` diff --git a/docs/dyn/displayvideo_v3.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v3.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html index e0441cb689..4bf2df7563 100644 --- a/docs/dyn/displayvideo_v3.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v3.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html @@ -138,7 +138,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -300,7 +299,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -321,7 +320,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -380,7 +379,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -641,7 +640,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -662,7 +661,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -721,7 +720,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -876,7 +875,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -949,7 +947,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1117,7 +1114,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1138,7 +1135,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1197,7 +1194,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1352,7 +1349,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1525,7 +1521,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1546,7 +1542,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1605,7 +1601,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v3.advertisers.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v3.advertisers.targetingTypes.assignedTargetingOptions.html index bc8fdb0227..74360bbe1f 100644 --- a/docs/dyn/displayvideo_v3.advertisers.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v3.advertisers.targetingTypes.assignedTargetingOptions.html @@ -137,7 +137,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -299,7 +298,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -320,7 +319,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -379,7 +378,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -640,7 +639,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -661,7 +660,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -720,7 +719,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -874,7 +873,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -946,7 +944,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1114,7 +1111,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1135,7 +1132,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1194,7 +1191,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1348,7 +1345,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1521,7 +1517,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1542,7 +1538,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1601,7 +1597,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v3.partners.html b/docs/dyn/displayvideo_v3.partners.html index aaf6994170..e1b56e9c68 100644 --- a/docs/dyn/displayvideo_v3.partners.html +++ b/docs/dyn/displayvideo_v3.partners.html @@ -257,7 +257,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -278,7 +278,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -337,7 +337,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -613,7 +613,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -634,7 +634,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -693,7 +693,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v3.partners.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v3.partners.targetingTypes.assignedTargetingOptions.html index 0d01db4831..a4546bfa30 100644 --- a/docs/dyn/displayvideo_v3.partners.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v3.partners.targetingTypes.assignedTargetingOptions.html @@ -137,7 +137,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -299,7 +298,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -320,7 +319,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -379,7 +378,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -640,7 +639,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -661,7 +660,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -720,7 +719,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -874,7 +873,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -946,7 +944,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1114,7 +1111,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1135,7 +1132,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1194,7 +1191,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1348,7 +1345,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -1521,7 +1517,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1542,7 +1538,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1601,7 +1597,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v3.targetingTypes.targetingOptions.html b/docs/dyn/displayvideo_v3.targetingTypes.targetingOptions.html index 3b1de218ad..eb40225754 100644 --- a/docs/dyn/displayvideo_v3.targetingTypes.targetingOptions.html +++ b/docs/dyn/displayvideo_v3.targetingTypes.targetingOptions.html @@ -136,7 +136,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -262,7 +261,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. @@ -319,7 +318,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -451,7 +449,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. @@ -524,7 +522,6 @@Method Details
TARGETING_TYPE_NEGATIVE_KEYWORD_LIST - Target ads to a specific negative keyword list. TARGETING_TYPE_VIEWABILITY - Target ads to a specific viewability (for example, 80% viewable). TARGETING_TYPE_CATEGORY - Target ads to a specific content category (for example, arts & entertainment). - TARGETING_TYPE_INVENTORY_SOURCE - Purchase impressions from specific deals and auction packages. TARGETING_TYPE_LANGUAGE - Target ads to a specific language (for example, English or Japanese). TARGETING_TYPE_AUTHORIZED_SELLER_STATUS - Target ads to ads.txt authorized sellers. If no targeting option of this type is assigned, the resource uses the "Authorized Direct Sellers and Resellers" option by default. TARGETING_TYPE_GEO_REGION - Target ads to a specific regional location (for example, a city or state). @@ -670,7 +667,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. diff --git a/docs/dyn/displayvideo_v4.advertisers.adGroupAds.html b/docs/dyn/displayvideo_v4.advertisers.adGroupAds.html index 698f3a5f20..71562d4cd9 100644 --- a/docs/dyn/displayvideo_v4.advertisers.adGroupAds.html +++ b/docs/dyn/displayvideo_v4.advertisers.adGroupAds.html @@ -279,7 +279,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -502,7 +502,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -552,7 +552,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -792,7 +792,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1015,7 +1015,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1065,7 +1065,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1332,7 +1332,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1555,7 +1555,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1605,7 +1605,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -1858,7 +1858,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2081,7 +2081,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2131,7 +2131,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2391,7 +2391,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2614,7 +2614,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2664,7 +2664,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -2905,7 +2905,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -3128,7 +3128,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. @@ -3178,7 +3178,7 @@Method Details
"displayUrl": "A String", # Required. The webpage address that appears with the ad. "finalUrl": "A String", # Required. The URL address of the webpage that people reach after they click the ad. "trackingUrl": "A String", # Output only. The URL address loaded in the background for tracking purposes. - "video": { # Details of a YouTube video. # Required. The YouTube video of the ad. + "video": { # Details of a YouTube video. # Required. Immutable. The YouTube video of the ad. "id": "A String", # Output only. The YouTube video ID which can be searched on YouTube webpage. "unavailableReason": "A String", # Output only. The reason why the video data is not available. "videoAssetId": "A String", # Required. The YouTube video asset id. This is the adAssetId of an AdAsset resource. diff --git a/docs/dyn/displayvideo_v4.advertisers.adGroups.html b/docs/dyn/displayvideo_v4.advertisers.adGroups.html index 911692a325..c228c02877 100644 --- a/docs/dyn/displayvideo_v4.advertisers.adGroups.html +++ b/docs/dyn/displayvideo_v4.advertisers.adGroups.html @@ -270,7 +270,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -291,7 +291,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -350,7 +350,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -668,7 +668,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -689,7 +689,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -748,7 +748,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v4.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html index 0b8d4a2d27..3f6cc79ec7 100644 --- a/docs/dyn/displayvideo_v4.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v4.advertisers.adGroups.targetingTypes.assignedTargetingOptions.html @@ -300,7 +300,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -321,7 +321,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -380,7 +380,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -641,7 +641,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -662,7 +662,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -721,7 +721,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1117,7 +1117,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1138,7 +1138,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1197,7 +1197,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1525,7 +1525,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1546,7 +1546,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1605,7 +1605,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.advertisers.html b/docs/dyn/displayvideo_v4.advertisers.html index 3a57a74c23..a0151e827f 100644 --- a/docs/dyn/displayvideo_v4.advertisers.html +++ b/docs/dyn/displayvideo_v4.advertisers.html @@ -519,7 +519,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -540,7 +540,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -599,7 +599,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -875,7 +875,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -896,7 +896,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -955,7 +955,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1392,7 +1392,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1413,7 +1413,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1472,7 +1472,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.advertisers.lineItems.html b/docs/dyn/displayvideo_v4.advertisers.lineItems.html index 45682e8a16..b2c5a511d2 100644 --- a/docs/dyn/displayvideo_v4.advertisers.lineItems.html +++ b/docs/dyn/displayvideo_v4.advertisers.lineItems.html @@ -273,7 +273,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -294,7 +294,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -353,7 +353,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -673,7 +673,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -694,7 +694,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -753,7 +753,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1052,7 +1052,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1067,8 +1067,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -1323,7 +1323,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1338,8 +1338,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -1554,7 +1554,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1569,8 +1569,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -1840,7 +1840,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -1855,8 +1855,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -2084,7 +2084,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -2099,8 +2099,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -2335,7 +2335,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -2350,8 +2350,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` @@ -2567,7 +2567,7 @@Method Details
"warningMessages": [ # Output only. The warning messages generated by the line item. These warnings do not block saving the line item, but some may block the line item from running. "A String", ], - "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Output only. Settings specific to YouTube and Partners line items. + "youtubeAndPartnersSettings": { # Settings for YouTube and Partners line items. # Optional. Settings specific to YouTube and Partners line items. "contentCategory": "A String", # Output only. The kind of content on which the YouTube and Partners ads will be shown. *Warning*: This field will be removed in the near future. Use effective_content_category instead. "effectiveContentCategory": "A String", # Output only. The content category which takes effect when serving the line item. When content category is set in both line item and advertiser, the stricter one will take effect when serving the line item. New line items will only inherit the advertiser level setting. "inventorySourceSettings": { # Settings that control what YouTube related inventories the YouTube and Partners line item will target. # Settings that control what YouTube and Partners inventories the line item will target. @@ -2582,8 +2582,8 @@Method Details
], "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times you want ads from this line item to show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "thirdPartyMeasurementConfigs": { # Settings that control what third-party vendors are measuring specific line item metrics. # Optional. The third-party measurement configs of the line item. "brandLiftVendorConfigs": [ # Optional. The third-party vendors measuring brand lift. The following third-party vendors are applicable: * `THIRD_PARTY_VENDOR_DYNATA` * `THIRD_PARTY_VENDOR_KANTAR` * `THIRD_PARTY_VENDOR_INTAGE` * `THIRD_PARTY_VENDOR_MACROMILL` diff --git a/docs/dyn/displayvideo_v4.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v4.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html index 3b96ea5c16..66e8955970 100644 --- a/docs/dyn/displayvideo_v4.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v4.advertisers.lineItems.targetingTypes.assignedTargetingOptions.html @@ -300,7 +300,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -321,7 +321,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -380,7 +380,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -641,7 +641,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -662,7 +662,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -721,7 +721,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1117,7 +1117,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1138,7 +1138,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1197,7 +1197,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1525,7 +1525,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1546,7 +1546,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1605,7 +1605,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.advertisers.reachForecast.html b/docs/dyn/displayvideo_v4.advertisers.reachForecast.html index fb1c32042e..25e8b247a4 100644 --- a/docs/dyn/displayvideo_v4.advertisers.reachForecast.html +++ b/docs/dyn/displayvideo_v4.advertisers.reachForecast.html @@ -168,8 +168,8 @@Method Details
}, "targetFrequency": { # Setting that controls the average number of times the ads will show to the same person over a certain period of time. # Optional. The average number of times the ads will show to the same person over a certain period of time. "targetCount": "A String", # The target number of times, on average, the ads will be shown to the same person in the timespan dictated by time_unit and time_unit_count. - "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` - "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 + "timeUnit": "A String", # The unit of time in which the target frequency will be applied. The following time unit is applicable: * `TIME_UNIT_WEEKS` * `TIME_UNIT_MONTHS` + "timeUnitCount": 42, # The number of time_unit the target frequency will last. The following restrictions apply based on the value of time_unit: * `TIME_UNIT_WEEKS` - must be 1 * `TIME_UNIT_MONTHS` - must be 1 }, "userInterestIds": [ # Optional. The user interest IDs to target. Plannable user interests can be retrieved using the `RetrievePlannableUserInterests` method. "A String", diff --git a/docs/dyn/displayvideo_v4.advertisers.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v4.advertisers.targetingTypes.assignedTargetingOptions.html index ea8860ade8..1432fc7e24 100644 --- a/docs/dyn/displayvideo_v4.advertisers.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v4.advertisers.targetingTypes.assignedTargetingOptions.html @@ -299,7 +299,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -320,7 +320,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -379,7 +379,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -640,7 +640,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -661,7 +661,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -720,7 +720,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1114,7 +1114,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1135,7 +1135,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1194,7 +1194,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1521,7 +1521,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1542,7 +1542,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1601,7 +1601,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.partners.html b/docs/dyn/displayvideo_v4.partners.html index 2872fea96c..8961167c9c 100644 --- a/docs/dyn/displayvideo_v4.partners.html +++ b/docs/dyn/displayvideo_v4.partners.html @@ -257,7 +257,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -278,7 +278,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -337,7 +337,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -613,7 +613,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -634,7 +634,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -693,7 +693,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.partners.targetingTypes.assignedTargetingOptions.html b/docs/dyn/displayvideo_v4.partners.targetingTypes.assignedTargetingOptions.html index 8c4458f132..2a56521c77 100644 --- a/docs/dyn/displayvideo_v4.partners.targetingTypes.assignedTargetingOptions.html +++ b/docs/dyn/displayvideo_v4.partners.targetingTypes.assignedTargetingOptions.html @@ -299,7 +299,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -320,7 +320,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -379,7 +379,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -640,7 +640,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -661,7 +661,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -720,7 +720,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1114,7 +1114,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1135,7 +1135,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1194,7 +1194,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. @@ -1521,7 +1521,7 @@Method Details
"youtubeAndPartnersBidMultiplier": 3.14, # Optional. Bid multiplier allows you to show your ads more or less frequently based on the device type. It will apply a multiplier on the original bid price. When this field is 0, it indicates this field is not applicable instead of multiplying 0 on the original bid price. For example, if the bid price without multiplier is $10.0 and the multiplier is 1.5 for Tablet, the resulting bid price for Tablet will be $15.0. Only applicable to YouTube and Partners line items. }, "digitalContentLabelExclusionDetails": { # Targeting details for digital content label. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. # Digital content label details. This field will be populated when the targeting_type is `TARGETING_TYPE_DIGITAL_CONTENT_LABEL_EXCLUSION`. Digital content labels are targeting exclusions. Advertiser level digital content label exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude content labels in addition to advertiser exclusions, but can't override them. A line item won't serve if all the digital content labels are excluded. - "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept the value `CONTENT_RATING_TIER_UNRATED`. All other values will be deprecated and no longer be accepted.** + "excludedContentRatingTier": "A String", # Required. The display name of the digital content label rating tier to be EXCLUDED. This field only accepts the value `CONTENT_RATING_TIER_UNRATED`. }, "environmentDetails": { # Assigned environment targeting option details. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_ENVIRONMENT`. # Environment details. This field will be populated when the targeting_type is `TARGETING_TYPE_ENVIRONMENT`. "environment": "A String", # Required. The serving environment. @@ -1542,7 +1542,7 @@Method Details
"householdIncome": "A String", # Required. The household income of the audience. }, "inheritance": "A String", # Output only. The inheritance status of the assigned targeting option. - "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE`. + "inventorySourceDetails": { # Targeting details for inventory source. This will be populated in the details field of an AssignedTargetingOption when targeting_type is one of `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. # Inventory source details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE` or `TARGETING_TYPE_INVENTORY_SOURCE_DEAL`. "inventorySourceId": "A String", # Required. ID of the inventory source. Should refer to the inventory_source_id field of an InventorySource resource. }, "inventorySourceGroupDetails": { # Targeting details for inventory source group. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. # Inventory source group details. This field will be populated when the targeting_type is `TARGETING_TYPE_INVENTORY_SOURCE_GROUP`. @@ -1601,7 +1601,7 @@Method Details
"regionalLocationListId": "A String", # Required. ID of the regional location list. Should refer to the location_list_id field of a LocationList resource whose type is `TARGETING_LOCATION_TYPE_REGIONAL`. }, "sensitiveCategoryExclusionDetails": { # Targeting details for sensitive category. This will be populated in the details field of an AssignedTargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive category details. This field will be populated when the targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. Sensitive categories are targeting exclusions. Advertiser level sensitive category exclusions, if set, are always applied in serving (even though they aren't visible in resource settings). Resource settings can exclude sensitive categories in addition to advertiser exclusions, but can't override them. - "excludedSensitiveCategory": "A String", # Required. An enum for the DV360 Sensitive category content classified to be EXCLUDED. **Starting on *October 1, 2026*, this field will only accept `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` or `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO`. All other values will be deprecated and no longer be accepted.** + "excludedSensitiveCategory": "A String", # Required. An enum for the Display & Video 360 Sensitive category content classified to be EXCLUDED. This field only accepts the following values: * `SENSITIVE_CATEGORY_EMBEDDED_VIDEO` * `SENSITIVE_CATEGORY_LIVE_STREAMING_VIDEO` }, "sessionPositionDetails": { # Details for session position assigned targeting option. This will be populated in the session_position_details field when targeting_type is `TARGETING_TYPE_SESSION_POSITION`. # Session position details. This field will be populated when the targeting_type is `TARGETING_TYPE_SESSION_POSITION`. "sessionPosition": "A String", # The position where the ad will show in a session. diff --git a/docs/dyn/displayvideo_v4.targetingTypes.targetingOptions.html b/docs/dyn/displayvideo_v4.targetingTypes.targetingOptions.html index f1bba765ac..b750587cda 100644 --- a/docs/dyn/displayvideo_v4.targetingTypes.targetingOptions.html +++ b/docs/dyn/displayvideo_v4.targetingTypes.targetingOptions.html @@ -262,7 +262,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. @@ -451,7 +451,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. @@ -670,7 +670,7 @@Method Details
"longitude": 3.14, # Output only. Longitude of the POI rounding to 6th decimal place. }, "sensitiveCategoryDetails": { # Represents a targetable sensitive category. This will be populated in the sensitive_category_details field of the TargetingOption when targeting_type is `TARGETING_TYPE_SENSITIVE_CATEGORY_EXCLUSION`. # Sensitive Category details. - "sensitiveCategory": "A String", # Output only. An enum for the DV360 Sensitive category content classifier. + "sensitiveCategory": "A String", # Output only. An enum for the Display & Video 360 Sensitive category content classifier. }, "subExchangeDetails": { # Represents a targetable sub-exchange. This will be populated in the sub_exchange_details field of a TargetingOption when targeting_type is `TARGETING_TYPE_SUB_EXCHANGE`. # Sub-exchange details. "displayName": "A String", # Output only. The display name of the sub-exchange. diff --git a/docs/dyn/dns_v1.html b/docs/dyn/dns_v1.html index ab4f33e804..38b0911e9c 100644 --- a/docs/dyn/dns_v1.html +++ b/docs/dyn/dns_v1.html @@ -84,6 +84,11 @@Instance Methods
Returns the dnsKeys Resource.
++
+locations()+Returns the locations Resource.
+ @@ -94,6 +99,16 @@Instance Methods
Returns the managedZones Resource.
++
+operations()+Returns the operations Resource.
+ + +Returns the outboundEndpoints Resource.
+ diff --git a/docs/dyn/dns_v1.locations.html b/docs/dyn/dns_v1.locations.html new file mode 100644 index 0000000000..cbf09e684e --- /dev/null +++ b/docs/dyn/dns_v1.locations.html @@ -0,0 +1,172 @@ + + + +Cloud DNS API . locations
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets information about a location.
++
+list(name, extraLocationTypes=None, filter=None, pageSize=None, pageToken=None, x__xgafv=None)Lists information about the supported locations for this service. This method lists locations based on the resource scope provided in the ListLocationsRequest.name field: * **Global locations**: If `name` is empty, the method lists the public locations available to all projects. * **Project-specific locations**: If `name` follows the format `projects/{project}`, the method lists locations visible to that specific project. This includes public, private, or other project-specific locations enabled for the project. For gRPC and client library implementations, the resource name is passed as the `name` field. For direct service calls, the resource name is incorporated into the request path based on the specific service implementation and version.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +get(name, x__xgafv=None)+Gets information about a location. + +Args: + name: string, Resource name for the location. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # A resource that represents a Google Cloud location. + "displayName": "A String", # The friendly name for this location, typically a nearby city name. For example, "Tokyo". + "labels": { # Cross-service attributes for the location. For example {"cloud.googleapis.com/region": "us-east1"} + "a_key": "A String", + }, + "locationId": "A String", # The canonical id for this location. For example: `"us-east1"`. + "metadata": { # Service-specific metadata. For example the available capacity at the given location. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # Resource name for the location, which may vary between implementations. For example: `"projects/example-project/locations/us-east1"` +}+++ +list(name, extraLocationTypes=None, filter=None, pageSize=None, pageToken=None, x__xgafv=None)+Lists information about the supported locations for this service. This method lists locations based on the resource scope provided in the ListLocationsRequest.name field: * **Global locations**: If `name` is empty, the method lists the public locations available to all projects. * **Project-specific locations**: If `name` follows the format `projects/{project}`, the method lists locations visible to that specific project. This includes public, private, or other project-specific locations enabled for the project. For gRPC and client library implementations, the resource name is passed as the `name` field. For direct service calls, the resource name is incorporated into the request path based on the specific service implementation and version. + +Args: + name: string, The resource that owns the locations collection, if applicable. (required) + extraLocationTypes: string, Optional. Do not use this field unless explicitly documented otherwise. This is primarily for internal usage. (repeated) + filter: string, A filter to narrow down results to a preferred subset. The filtering language accepts strings like `"displayName=tokyo"`, and is documented in more detail in [AIP-160](https://google.aip.dev/160). + pageSize: integer, The maximum number of results to return. If not set, the service selects a default. + pageToken: string, A page token received from the `next_page_token` field in the response. Send that page token to receive the subsequent page. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # The response message for Locations.ListLocations. + "locations": [ # A list of locations that matches the specified filter in the request. + { # A resource that represents a Google Cloud location. + "displayName": "A String", # The friendly name for this location, typically a nearby city name. For example, "Tokyo". + "labels": { # Cross-service attributes for the location. For example {"cloud.googleapis.com/region": "us-east1"} + "a_key": "A String", + }, + "locationId": "A String", # The canonical id for this location. For example: `"us-east1"`. + "metadata": { # Service-specific metadata. For example the available capacity at the given location. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # Resource name for the location, which may vary between implementations. For example: `"projects/example-project/locations/us-east1"` + }, + ], + "nextPageToken": "A String", # The standard List next-page token. +}+++ + \ No newline at end of file diff --git a/docs/dyn/dns_v1.managedZoneOperations.html b/docs/dyn/dns_v1.managedZoneOperations.html index 2917833a92..ddeccc01eb 100644 --- a/docs/dyn/dns_v1.managedZoneOperations.html +++ b/docs/dyn/dns_v1.managedZoneOperations.html @@ -178,6 +178,12 @@list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -257,6 +263,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -411,6 +423,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -490,6 +508,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. diff --git a/docs/dyn/dns_v1.managedZones.html b/docs/dyn/dns_v1.managedZones.html index 85ee981973..4507dcdcf0 100644 --- a/docs/dyn/dns_v1.managedZones.html +++ b/docs/dyn/dns_v1.managedZones.html @@ -145,6 +145,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -234,6 +240,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -346,6 +358,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -503,6 +521,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -611,6 +635,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -746,6 +776,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -825,6 +861,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -1033,6 +1075,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -1168,6 +1216,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -1247,6 +1301,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. diff --git a/docs/dyn/dns_v1.operations.html b/docs/dyn/dns_v1.operations.html new file mode 100644 index 0000000000..642faa96e3 --- /dev/null +++ b/docs/dyn/dns_v1.operations.html @@ -0,0 +1,233 @@ + + + +Cloud DNS API . operations
+Instance Methods
+ +Starts asynchronous cancellation on a long-running operation. The server makes a best effort to cancel the operation, but success is not guaranteed. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. Clients can use Operations.GetOperation or other methods to check whether the cancellation succeeded or whether the operation completed despite cancellation. On successful cancellation, the operation is not deleted; instead, it becomes an operation with an Operation.error value with a google.rpc.Status.code of `1`, corresponding to `Code.CANCELLED`.
++
+close()Close httplib2 connections.
+ +Deletes a long-running operation. This method indicates that the client is no longer interested in the operation result. It does not cancel the operation. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`.
+ +Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service.
++
+list(name, filter=None, pageSize=None, pageToken=None, returnPartialSuccess=None, x__xgafv=None)Lists operations that match the specified filter in the request. If the server doesn't support this method, it returns `UNIMPLEMENTED`.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +cancel(name, x__xgafv=None)+Starts asynchronous cancellation on a long-running operation. The server makes a best effort to cancel the operation, but success is not guaranteed. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. Clients can use Operations.GetOperation or other methods to check whether the cancellation succeeded or whether the operation completed despite cancellation. On successful cancellation, the operation is not deleted; instead, it becomes an operation with an Operation.error value with a google.rpc.Status.code of `1`, corresponding to `Code.CANCELLED`. + +Args: + name: string, The name of the operation resource to be cancelled. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); } +}+++ +close()+Close httplib2 connections.+++ +delete(name, x__xgafv=None)+Deletes a long-running operation. This method indicates that the client is no longer interested in the operation result. It does not cancel the operation. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. + +Args: + name: string, The name of the operation resource to be deleted. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); } +}+++ +get(name, x__xgafv=None)+Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service. + +Args: + name: string, The name of the operation resource. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+++ +list(name, filter=None, pageSize=None, pageToken=None, returnPartialSuccess=None, x__xgafv=None)+Lists operations that match the specified filter in the request. If the server doesn't support this method, it returns `UNIMPLEMENTED`. + +Args: + name: string, The name of the operation's parent resource. (required) + filter: string, The standard list filter. + pageSize: integer, The standard list page size. + pageToken: string, The standard list page token. + returnPartialSuccess: boolean, When set to `true`, operations that are reachable are returned as normal, and those that are unreachable are returned in the ListOperationsResponse.unreachable field. This can only be `true` when reading across collections. For example, when `parent` is set to `"projects/example/locations/-"`. This field is not supported by default and will result in an `UNIMPLEMENTED` error if set unless explicitly documented otherwise in service or product specific documentation. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # The response message for Operations.ListOperations. + "nextPageToken": "A String", # The standard List next-page token. + "operations": [ # A list of operations that matches the specified filter in the request. + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + }, + ], + "unreachable": [ # Unordered list. Unreachable resources. Populated when the request sets `ListOperationsRequest.return_partial_success` and reads across collections. For example, when attempting to list all resources across all supported locations. + "A String", + ], +}+++ + \ No newline at end of file diff --git a/docs/dyn/dns_v1.outboundEndpoints.html b/docs/dyn/dns_v1.outboundEndpoints.html new file mode 100644 index 0000000000..1c92da74cd --- /dev/null +++ b/docs/dyn/dns_v1.outboundEndpoints.html @@ -0,0 +1,355 @@ + + + +list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Cloud DNS API . outboundEndpoints
+Instance Methods
++
+close()Close httplib2 connections.
+ +Creates a new Outbound Endpoint.
++
+delete(name, clientOperationId=None, requestId=None, x__xgafv=None)Deletes a previously created Outbound Endpoint.
++
+get(name, clientOperationId=None, x__xgafv=None)Fetches the representation of an existing Outbound Endpoint.
++
+list(parent, clientOperationId=None, pageSize=None, pageToken=None, x__xgafv=None)Enumerates all Outbound Endpoints associated with a project.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
++
+patch(name, body=None, clientOperationId=None, requestId=None, updateMask=None, x__xgafv=None)Updates an existing Outbound Endpoint.
+Method Details
+++ +close()+Close httplib2 connections.+++ +create(parent, body=None, clientOperationId=None, outboundEndpointId=None, requestId=None, x__xgafv=None)+Creates a new Outbound Endpoint. + +Args: + parent: string, Required. The parent project and location where this OutboundEndpoint will be created. Format: projects/{project}/locations/{location} (required) + body: object, The request body. + The object takes the form of: + +{ # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp +} + + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + outboundEndpointId: string, Required. The ID to use for the OutboundEndpoint, which will become the final component of the OutboundEndpoint's resource name. + requestId: string, Optional. An optional request ID to identify requests. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+++ +delete(name, clientOperationId=None, requestId=None, x__xgafv=None)+Deletes a previously created Outbound Endpoint. + +Args: + name: string, Required. The name of the OutboundEndpoint to delete. Format: projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint} (required) + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + requestId: string, Optional. An optional request ID to identify requests. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+++ +get(name, clientOperationId=None, x__xgafv=None)+Fetches the representation of an existing Outbound Endpoint. + +Args: + name: string, Required. The name of the OutboundEndpoint to retrieve. Format: projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint} (required) + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp +}+++ +list(parent, clientOperationId=None, pageSize=None, pageToken=None, x__xgafv=None)+Enumerates all Outbound Endpoints associated with a project. + +Args: + parent: string, Required. The parent project and location from which to list resources. Format: projects/{project}/locations/{location} (required) + clientOperationId: string, A parameter + pageSize: integer, Optional. The maximum number of results to return. + pageToken: string, Optional. A page token received from a previous List call. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Response message for OutboundEndpointsService.ListOutboundEndpoints + "nextPageToken": "A String", # A token to retrieve the next page of results. Set to empty if there are no remaining results. + "outboundEndpoints": [ # The list of OutboundEndpoints. + { # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp + }, + ], + "unreachable": [ # Unordered list. The resource names of the unreachable locations. Format: `projects/{project}/locations/{location}` + "A String", + ], +}+++ +list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++++ + \ No newline at end of file diff --git a/docs/dyn/dns_v1.policies.html b/docs/dyn/dns_v1.policies.html index d2098776cd..91ff8034c8 100644 --- a/docs/dyn/dns_v1.policies.html +++ b/docs/dyn/dns_v1.policies.html @@ -116,6 +116,12 @@patch(name, body=None, clientOperationId=None, requestId=None, updateMask=None, x__xgafv=None)+Updates an existing Outbound Endpoint. + +Args: + name: string, Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` (required) + body: object, The request body. + The object takes the form of: + +{ # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp +} + + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + requestId: string, Optional. An optional request ID to identify requests. + updateMask: string, Required. The list of fields to be updated. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -158,6 +164,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -223,6 +235,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -277,6 +295,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -337,6 +361,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -380,6 +410,12 @@Method Details
"policy": { # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -425,6 +461,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -468,6 +510,12 @@Method Details
"policy": { # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. diff --git a/docs/dyn/dns_v1.projects.html b/docs/dyn/dns_v1.projects.html index fd65e80559..f69c849276 100644 --- a/docs/dyn/dns_v1.projects.html +++ b/docs/dyn/dns_v1.projects.html @@ -119,6 +119,8 @@Method Details
"networksPerManagedZone": 42, # Maximum allowed number of networks to which a privately scoped zone can be attached. "networksPerPolicy": 42, # Maximum allowed number of networks per policy. "networksPerResponsePolicy": 42, # Maximum allowed number of networks per response policy. + "outboundEndpointsPerManagedZone": 42, # Maximum allowed number of outbound endpoints per managed zone. + "outboundEndpointsPerPolicy": 42, # Maximum allowed number of outbound endpoints per policy. "peeringZonesPerTargetNetwork": 42, # Maximum allowed number of consumer peering zones per target network owned by this producer project "policies": 42, # Maximum allowed number of policies per project. "resourceRecordsPerRrset": 42, # Maximum allowed number of ResourceRecords per ResourceRecordSet. diff --git a/docs/dyn/dns_v1beta2.html b/docs/dyn/dns_v1beta2.html index 5c4842d6d3..37ee2f84d4 100644 --- a/docs/dyn/dns_v1beta2.html +++ b/docs/dyn/dns_v1beta2.html @@ -84,6 +84,11 @@Instance Methods
Returns the dnsKeys Resource.
++
+locations()+Returns the locations Resource.
+ @@ -94,6 +99,16 @@Instance Methods
Returns the managedZones Resource.
++
+operations()+Returns the operations Resource.
+ + +Returns the outboundEndpoints Resource.
+ diff --git a/docs/dyn/dns_v1beta2.locations.html b/docs/dyn/dns_v1beta2.locations.html new file mode 100644 index 0000000000..97571da216 --- /dev/null +++ b/docs/dyn/dns_v1beta2.locations.html @@ -0,0 +1,172 @@ + + + +Cloud DNS API . locations
+Instance Methods
++
+close()Close httplib2 connections.
+ +Gets information about a location.
++
+list(name, extraLocationTypes=None, filter=None, pageSize=None, pageToken=None, x__xgafv=None)Lists information about the supported locations for this service. This method lists locations based on the resource scope provided in the ListLocationsRequest.name field: * **Global locations**: If `name` is empty, the method lists the public locations available to all projects. * **Project-specific locations**: If `name` follows the format `projects/{project}`, the method lists locations visible to that specific project. This includes public, private, or other project-specific locations enabled for the project. For gRPC and client library implementations, the resource name is passed as the `name` field. For direct service calls, the resource name is incorporated into the request path based on the specific service implementation and version.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +close()+Close httplib2 connections.+++ +get(name, x__xgafv=None)+Gets information about a location. + +Args: + name: string, Resource name for the location. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # A resource that represents a Google Cloud location. + "displayName": "A String", # The friendly name for this location, typically a nearby city name. For example, "Tokyo". + "labels": { # Cross-service attributes for the location. For example {"cloud.googleapis.com/region": "us-east1"} + "a_key": "A String", + }, + "locationId": "A String", # The canonical id for this location. For example: `"us-east1"`. + "metadata": { # Service-specific metadata. For example the available capacity at the given location. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # Resource name for the location, which may vary between implementations. For example: `"projects/example-project/locations/us-east1"` +}+++ +list(name, extraLocationTypes=None, filter=None, pageSize=None, pageToken=None, x__xgafv=None)+Lists information about the supported locations for this service. This method lists locations based on the resource scope provided in the ListLocationsRequest.name field: * **Global locations**: If `name` is empty, the method lists the public locations available to all projects. * **Project-specific locations**: If `name` follows the format `projects/{project}`, the method lists locations visible to that specific project. This includes public, private, or other project-specific locations enabled for the project. For gRPC and client library implementations, the resource name is passed as the `name` field. For direct service calls, the resource name is incorporated into the request path based on the specific service implementation and version. + +Args: + name: string, The resource that owns the locations collection, if applicable. (required) + extraLocationTypes: string, Optional. Do not use this field unless explicitly documented otherwise. This is primarily for internal usage. (repeated) + filter: string, A filter to narrow down results to a preferred subset. The filtering language accepts strings like `"displayName=tokyo"`, and is documented in more detail in [AIP-160](https://google.aip.dev/160). + pageSize: integer, The maximum number of results to return. If not set, the service selects a default. + pageToken: string, A page token received from the `next_page_token` field in the response. Send that page token to receive the subsequent page. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # The response message for Locations.ListLocations. + "locations": [ # A list of locations that matches the specified filter in the request. + { # A resource that represents a Google Cloud location. + "displayName": "A String", # The friendly name for this location, typically a nearby city name. For example, "Tokyo". + "labels": { # Cross-service attributes for the location. For example {"cloud.googleapis.com/region": "us-east1"} + "a_key": "A String", + }, + "locationId": "A String", # The canonical id for this location. For example: `"us-east1"`. + "metadata": { # Service-specific metadata. For example the available capacity at the given location. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # Resource name for the location, which may vary between implementations. For example: `"projects/example-project/locations/us-east1"` + }, + ], + "nextPageToken": "A String", # The standard List next-page token. +}+++ + \ No newline at end of file diff --git a/docs/dyn/dns_v1beta2.managedZoneOperations.html b/docs/dyn/dns_v1beta2.managedZoneOperations.html index ddee9f0bc1..6a4f797cfb 100644 --- a/docs/dyn/dns_v1beta2.managedZoneOperations.html +++ b/docs/dyn/dns_v1beta2.managedZoneOperations.html @@ -178,6 +178,12 @@list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -257,6 +263,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -411,6 +423,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -490,6 +508,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. diff --git a/docs/dyn/dns_v1beta2.managedZones.html b/docs/dyn/dns_v1beta2.managedZones.html index 9053db83d0..4cfab0d3c8 100644 --- a/docs/dyn/dns_v1beta2.managedZones.html +++ b/docs/dyn/dns_v1beta2.managedZones.html @@ -145,6 +145,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -234,6 +240,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -346,6 +358,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -503,6 +521,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -611,6 +635,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -746,6 +776,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -825,6 +861,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -1033,6 +1075,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -1168,6 +1216,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. @@ -1247,6 +1301,12 @@Method Details
}, "forwardingConfig": { # The presence for this field indicates that outbound forwarding is enabled for this zone. The value of this field contains the set of destinations to forward to. "kind": "dns#managedZoneForwardingConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#managedZoneForwardingConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # List of target name servers to forward to. Cloud DNS selects the best available name server if more than one target is given. { "domainName": "A String", # Fully qualified domain name for the forwarding target. diff --git a/docs/dyn/dns_v1beta2.operations.html b/docs/dyn/dns_v1beta2.operations.html new file mode 100644 index 0000000000..664b7aec02 --- /dev/null +++ b/docs/dyn/dns_v1beta2.operations.html @@ -0,0 +1,233 @@ + + + +Cloud DNS API . operations
+Instance Methods
+ +Starts asynchronous cancellation on a long-running operation. The server makes a best effort to cancel the operation, but success is not guaranteed. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. Clients can use Operations.GetOperation or other methods to check whether the cancellation succeeded or whether the operation completed despite cancellation. On successful cancellation, the operation is not deleted; instead, it becomes an operation with an Operation.error value with a google.rpc.Status.code of `1`, corresponding to `Code.CANCELLED`.
++
+close()Close httplib2 connections.
+ +Deletes a long-running operation. This method indicates that the client is no longer interested in the operation result. It does not cancel the operation. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`.
+ +Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service.
++
+list(name, filter=None, pageSize=None, pageToken=None, returnPartialSuccess=None, x__xgafv=None)Lists operations that match the specified filter in the request. If the server doesn't support this method, it returns `UNIMPLEMENTED`.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
+Method Details
+++ +cancel(name, x__xgafv=None)+Starts asynchronous cancellation on a long-running operation. The server makes a best effort to cancel the operation, but success is not guaranteed. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. Clients can use Operations.GetOperation or other methods to check whether the cancellation succeeded or whether the operation completed despite cancellation. On successful cancellation, the operation is not deleted; instead, it becomes an operation with an Operation.error value with a google.rpc.Status.code of `1`, corresponding to `Code.CANCELLED`. + +Args: + name: string, The name of the operation resource to be cancelled. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); } +}+++ +close()+Close httplib2 connections.+++ +delete(name, x__xgafv=None)+Deletes a long-running operation. This method indicates that the client is no longer interested in the operation result. It does not cancel the operation. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. + +Args: + name: string, The name of the operation resource to be deleted. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); } +}+++ +get(name, x__xgafv=None)+Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service. + +Args: + name: string, The name of the operation resource. (required) + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+++ +list(name, filter=None, pageSize=None, pageToken=None, returnPartialSuccess=None, x__xgafv=None)+Lists operations that match the specified filter in the request. If the server doesn't support this method, it returns `UNIMPLEMENTED`. + +Args: + name: string, The name of the operation's parent resource. (required) + filter: string, The standard list filter. + pageSize: integer, The standard list page size. + pageToken: string, The standard list page token. + returnPartialSuccess: boolean, When set to `true`, operations that are reachable are returned as normal, and those that are unreachable are returned in the ListOperationsResponse.unreachable field. This can only be `true` when reading across collections. For example, when `parent` is set to `"projects/example/locations/-"`. This field is not supported by default and will result in an `UNIMPLEMENTED` error if set unless explicitly documented otherwise in service or product specific documentation. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # The response message for Operations.ListOperations. + "nextPageToken": "A String", # The standard List next-page token. + "operations": [ # A list of operations that matches the specified filter in the request. + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + }, + ], + "unreachable": [ # Unordered list. Unreachable resources. Populated when the request sets `ListOperationsRequest.return_partial_success` and reads across collections. For example, when attempting to list all resources across all supported locations. + "A String", + ], +}+++ + \ No newline at end of file diff --git a/docs/dyn/dns_v1beta2.outboundEndpoints.html b/docs/dyn/dns_v1beta2.outboundEndpoints.html new file mode 100644 index 0000000000..4324275bbc --- /dev/null +++ b/docs/dyn/dns_v1beta2.outboundEndpoints.html @@ -0,0 +1,355 @@ + + + +list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++Cloud DNS API . outboundEndpoints
+Instance Methods
++
+close()Close httplib2 connections.
+ +Creates a new Outbound Endpoint.
++
+delete(name, clientOperationId=None, requestId=None, x__xgafv=None)Deletes a previously created Outbound Endpoint.
++
+get(name, clientOperationId=None, x__xgafv=None)Fetches the representation of an existing Outbound Endpoint.
++
+list(parent, clientOperationId=None, pageSize=None, pageToken=None, x__xgafv=None)Enumerates all Outbound Endpoints associated with a project.
++
+list_next(previous_request, previous_response)Retrieves the next page of results.
++
+patch(name, body=None, clientOperationId=None, requestId=None, updateMask=None, x__xgafv=None)Updates an existing Outbound Endpoint.
+Method Details
+++ +close()+Close httplib2 connections.+++ +create(parent, body=None, clientOperationId=None, outboundEndpointId=None, requestId=None, x__xgafv=None)+Creates a new Outbound Endpoint. + +Args: + parent: string, Required. The parent project and location where this OutboundEndpoint will be created. Format: projects/{project}/locations/{location} (required) + body: object, The request body. + The object takes the form of: + +{ # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp +} + + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + outboundEndpointId: string, Required. The ID to use for the OutboundEndpoint, which will become the final component of the OutboundEndpoint's resource name. + requestId: string, Optional. An optional request ID to identify requests. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+++ +delete(name, clientOperationId=None, requestId=None, x__xgafv=None)+Deletes a previously created Outbound Endpoint. + +Args: + name: string, Required. The name of the OutboundEndpoint to delete. Format: projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint} (required) + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + requestId: string, Optional. An optional request ID to identify requests. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+++ +get(name, clientOperationId=None, x__xgafv=None)+Fetches the representation of an existing Outbound Endpoint. + +Args: + name: string, Required. The name of the OutboundEndpoint to retrieve. Format: projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint} (required) + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp +}+++ +list(parent, clientOperationId=None, pageSize=None, pageToken=None, x__xgafv=None)+Enumerates all Outbound Endpoints associated with a project. + +Args: + parent: string, Required. The parent project and location from which to list resources. Format: projects/{project}/locations/{location} (required) + clientOperationId: string, A parameter + pageSize: integer, Optional. The maximum number of results to return. + pageToken: string, Optional. A page token received from a previous List call. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # Response message for OutboundEndpointsService.ListOutboundEndpoints + "nextPageToken": "A String", # A token to retrieve the next page of results. Set to empty if there are no remaining results. + "outboundEndpoints": [ # The list of OutboundEndpoints. + { # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp + }, + ], + "unreachable": [ # Unordered list. The resource names of the unreachable locations. Format: `projects/{project}/locations/{location}` + "A String", + ], +}+++ +list_next(previous_request, previous_response)+Retrieves the next page of results. + +Args: + previous_request: The request for the previous page. (required) + previous_response: The response from the request for the previous page. (required) + +Returns: + A request object that you can call 'execute()' on to request the next + page. Returns None if there are no more items in the collection. ++++ + \ No newline at end of file diff --git a/docs/dyn/dns_v1beta2.policies.html b/docs/dyn/dns_v1beta2.policies.html index 693ce8a653..9e4cfff7ba 100644 --- a/docs/dyn/dns_v1beta2.policies.html +++ b/docs/dyn/dns_v1beta2.policies.html @@ -116,6 +116,12 @@patch(name, body=None, clientOperationId=None, requestId=None, updateMask=None, x__xgafv=None)+Updates an existing Outbound Endpoint. + +Args: + name: string, Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` (required) + body: object, The request body. + The object takes the form of: + +{ # An OutboundEndpoint is a regional resource to enable outbound forwarding from a private IP range in VPCs to on-premise or other networks. + "annotations": { # Optional. Annotations as key value pairs + "a_key": "A String", + }, + "createTime": "A String", # Output only. [Output only] Create time stamp + "description": "A String", # Optional. User provided description. + "endpointIp": "A String", # Required. Immutable. User provided IP address to use for outbound DNS forwarding. This address must be a free address on the subnetwork. + "labels": { # Optional. Labels as key value pairs + "a_key": "A String", + }, + "name": "A String", # Identifier. The resource name of the OutboundEndpoint. Format: `projects/{project}/locations/{location}/outboundEndpoints/{outboundEndpoint}` + "network": "A String", # Required. Immutable. The VPC network containing the outbound endpoint, specified as full path. + "subnetwork": "A String", # Required. Immutable. The subnetwork holding the endpoint_ip, specified as full path. + "tags": { # Optional. Tag bindings as key value pairs + "a_key": "A String", + }, + "updateTime": "A String", # Output only. [Output only] Update time stamp +} + + clientOperationId: string, For mutating operation requests only. An optional identifier specified by the client. Must be unique for operation resources in the Operations collection. + requestId: string, Optional. An optional request ID to identify requests. + updateMask: string, Required. The list of fields to be updated. + x__xgafv: string, V1 error format. + Allowed values + 1 - v1 error format + 2 - v2 error format + +Returns: + An object of the form: + + { # This resource represents a long-running operation that is the result of a network API call. + "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available. + "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation. + "code": 42, # The status code, which should be an enum value of google.rpc.Code. + "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use. + { + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + ], + "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. + }, + "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, + "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`. + "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`. + "a_key": "", # Properties of the object. Contains field @type with type URL. + }, +}+Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -158,6 +164,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -223,6 +235,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -277,6 +295,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -337,6 +361,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -380,6 +410,12 @@Method Details
"policy": { # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -425,6 +461,12 @@Method Details
{ # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. @@ -468,6 +510,12 @@Method Details
"policy": { # A policy is a collection of DNS rules applied to one or more Virtual Private Cloud resources. "alternativeNameServerConfig": { # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. "kind": "dns#policyAlternativeNameServerConfig", + "outboundEndpoints": [ # The list of outbound endpoints to use for queries. + { + "kind": "dns#policyAlternativeNameServerConfigOutboundEndpoint", + "name": "A String", # Specified as a resource path including project and region. This should be formatted like projects/{project}/locations/{location}/outboundEndpoints/{endpoint} + }, + ], "targetNameServers": [ # Sets an alternative name server for the associated networks. When specified, all DNS queries are forwarded to a name server that you choose. Names such as .internal are not available when an alternative name server is specified. { "forwardingPath": "A String", # Forwarding path for this TargetNameServer. If unset or set to DEFAULT, Cloud DNS makes forwarding decisions based on address ranges; that is, RFC1918 addresses go to the VPC network, non-RFC1918 addresses go to the internet. When set to PRIVATE, Cloud DNS always sends queries through the VPC network for this target. diff --git a/docs/dyn/dns_v1beta2.projects.html b/docs/dyn/dns_v1beta2.projects.html index 1e64b6a974..52acaae50e 100644 --- a/docs/dyn/dns_v1beta2.projects.html +++ b/docs/dyn/dns_v1beta2.projects.html @@ -119,6 +119,8 @@Method Details
"networksPerManagedZone": 42, # Maximum allowed number of networks to which a privately scoped zone can be attached. "networksPerPolicy": 42, # Maximum allowed number of networks per policy. "networksPerResponsePolicy": 42, # Maximum allowed number of networks per response policy. + "outboundEndpointsPerManagedZone": 42, # Maximum allowed number of outbound endpoints per managed zone. + "outboundEndpointsPerPolicy": 42, # Maximum allowed number of outbound endpoints per policy. "peeringZonesPerTargetNetwork": 42, # Maximum allowed number of consumer peering zones per target network owned by this producer project "policies": 42, # Maximum allowed number of policies per project. "resourceRecordsPerRrset": 42, # Maximum allowed number of ResourceRecords per ResourceRecordSet. diff --git a/docs/dyn/docs_v1.documents.html b/docs/dyn/docs_v1.documents.html index 962014abc8..9e1f8cb5f1 100644 --- a/docs/dyn/docs_v1.documents.html +++ b/docs/dyn/docs_v1.documents.html @@ -99,14 +99,14 @@Method Details
{ # Request message for BatchUpdateDocument. "requests": [ # A list of updates to apply to the document. { # A single update to apply to a document. - "acceptSuggestion": { # Accepts a suggestion. Returns a 403 forbidden error if the requesting user does not have edit access to the document. [Developer Preview](https://developers.google.com/workspace/preview). # Accepts a suggestion. [Developer Preview](https://developers.google.com/workspace/preview). + "acceptSuggestion": { # Accepts a suggestion. Returns a 403 forbidden error if the requesting user does not have edit access to the document. # Accepts a suggestion. "suggestionId": "A String", # The ID of the suggestion. }, - "addCommentReply": { # Inserts a reply Post into a CommentThread or SuggestionThread. [Developer Preview](https://developers.google.com/workspace/preview). # Adds a reply to a CommentThread or SuggestionThread. [Developer Preview](https://developers.google.com/workspace/preview). + "addCommentReply": { # Inserts a reply Post into a CommentThread or SuggestionThread. # Adds a reply to a CommentThread or SuggestionThread. "commentId": "A String", # The ID of the CommentThread to add the reply to. - "post": { # Represents a single post in a comment or suggestion thread. [Developer Preview](https://developers.google.com/workspace/preview). # The Post representing the reply. + "post": { # Represents a single post in a comment or suggestion thread. # The Post representing the reply. "assigneeEmail": "A String", # Optional. The email of the user who is being newly assigned to the thread as part of this post. Returns a 400 bad request error if: - The parent thread is a CommentThread whose headPost does not have an assignee. - The parent thread is a SuggestionThread. - commentAction is specified as `RESOLVE` or `REOPEN`. - `assigneeEmail` exceeds 2048 UTF-8 code units. - "author": { # Represents a user who authored a comment or suggestion post. [Developer Preview](https://developers.google.com/workspace/preview). # Output only. The user who created the post. + "author": { # Represents a user who authored a comment or suggestion post. # Output only. The user who created the post. "anonymous": True or False, # Whether the user is anonymous. "displayName": "A String", # The display name of the user. May be absent if the author is anonymous. "me": True or False, # Whether the user is the authenticated user making the request. @@ -311,10 +311,10 @@Method Details
"tabId": "A String", # The tab that contains this range. When omitted, the request applies to the first tab. In a document containing a single tab: - If provided, must match the singular tab's ID. - If omitted, the request applies to the singular tab. In a document containing multiple tabs: - If provided, the request applies to the specified tab. - If omitted, the request applies to the first tab in the document. }, }, - "deleteComment": { # Deletes a CommentThread. Returns a 400 bad request error if the requesting user is not the author of the headPost. [Developer Preview](https://developers.google.com/workspace/preview). # Deletes a CommentThread. [Developer Preview](https://developers.google.com/workspace/preview). + "deleteComment": { # Deletes a CommentThread. Returns a 400 bad request error if the requesting user is not the author of the headPost. # Deletes a CommentThread. "commentId": "A String", # The ID of the CommentThread that is being deleted. }, - "deleteCommentReply": { # Deletes a reply Post from a CommentThread or SuggestionThread. Returns a 400 bad request error if: - The requesting user is not the author of the post. - The reply post contains an action. - The reply post contains an assignee. [Developer Preview](https://developers.google.com/workspace/preview). # Deletes a reply Post from a CommentThread or SuggestionThread. [Developer Preview](https://developers.google.com/workspace/preview). + "deleteCommentReply": { # Deletes a reply Post from a CommentThread or SuggestionThread. Returns a 400 bad request error if: - The requesting user is not the author of the post. - The reply post contains an action. - The reply post contains an assignee. # Deletes a reply Post from a CommentThread or SuggestionThread. "commentId": "A String", # The ID of the CommentThread which the post belongs to. "postId": "A String", # The ID of the reply Post being deleted. "suggestionId": "A String", # The ID of the SuggestionThread which the post belongs to. @@ -360,7 +360,7 @@Method Details
"objectId": "A String", # The ID of the positioned object to delete. "tabId": "A String", # The tab that the positioned object to delete is in. When omitted, the request is applied to the first tab. In a document containing a single tab: - If provided, must match the singular tab's ID. - If omitted, the request applies to the singular tab. In a document containing multiple tabs: - If provided, the request applies to the specified tab. - If omitted, the request applies to the first tab in the document. }, - "deleteSuggestion": { # Deletes a suggestion. Returns a 403 forbidden error if the requesting user is not the author of the suggestion. [Developer Preview](https://developers.google.com/workspace/preview). # Deletes a suggestion. [Developer Preview](https://developers.google.com/workspace/preview). + "deleteSuggestion": { # Deletes a suggestion. Returns a 403 forbidden error if the requesting user is not the author of the suggestion. # Deletes a suggestion. "suggestionId": "A String", # The ID of the suggestion. }, "deleteTab": { # Deletes a tab. If the tab has child tabs, they are deleted as well. # Deletes a document tab. @@ -388,7 +388,7 @@Method Details
}, }, }, - "insertComment": { # Inserts a CommentThread into the document. [Developer Preview](https://developers.google.com/workspace/preview). # Inserts a CommentThread into the document. [Developer Preview](https://developers.google.com/workspace/preview). + "insertComment": { # Inserts a CommentThread into the document. # Inserts a CommentThread into the document. "assigneeEmailAddress": "A String", # Optional. The email address of the assignee of the comment. Leave empty for a non-assigned comment. May not exceed 2048 UTF-8 code units. "content": "A String", # The text of the comment, as plain text. This text content will be handled similarly to comments created in the Docs editor. It will have similar behaviors for formatting, notifications, etc. This field cannot be empty, and must not exceed 2048 UTF-8 code units. "range": { # Specifies a contiguous range of text. # The Range in the document that is tied to this comment. @@ -578,7 +578,7 @@Method Details
"tabId": "A String", # The tab that the location is in. When omitted, the request is applied to the first tab. In a document containing a single tab: - If provided, must match the singular tab's ID. - If omitted, the request applies to the singular tab. In a document containing multiple tabs: - If provided, the request applies to the specified tab. - If omitted, the request applies to the first tab in the document. }, }, - "rejectSuggestion": { # Rejects a suggestion. Returns a 403 forbidden error if the requesting user does not have edit access to the document and is not the author of the suggestion. [Developer Preview](https://developers.google.com/workspace/preview). # Rejects a suggestion. [Developer Preview](https://developers.google.com/workspace/preview). + "rejectSuggestion": { # Rejects a suggestion. Returns a 403 forbidden error if the requesting user does not have edit access to the document and is not the author of the suggestion. # Rejects a suggestion. "suggestionId": "A String", # The ID of the suggestion. }, "replaceAllText": { # Replaces all instances of text matching a criteria with replace text. # Replaces all instances of the specified text. @@ -625,7 +625,7 @@Method Details
}, }, }, - "updateCommentPost": { # Updates a Post in a CommentThread or SuggestionThread. Returns a 400 bad request error if: - The post is the headPost of a SuggestionThread. - The requesting user is not the author of the post. [Developer Preview](https://developers.google.com/workspace/preview). # Updates an existing post (head post or reply) of a CommentThread or SuggestionThread. [Developer Preview](https://developers.google.com/workspace/preview). + "updateCommentPost": { # Updates a Post in a CommentThread or SuggestionThread. Returns a 400 bad request error if: - The post is the headPost of a SuggestionThread. - The requesting user is not the author of the post. # Updates an existing post (head post or reply) of a CommentThread or SuggestionThread. "commentId": "A String", # The ID of the CommentThread which the post belongs to. "content": "A String", # The new text of the comment, as plain text. This text content will be handled similarly to comments created in the Docs editor. It will have similar behaviors for formatting, notifications, etc. This field cannot be empty, and must not exceed 2048 UTF-8 code units. "postId": "A String", # The ID of the post being updated. @@ -1423,7 +1423,7 @@Method Details
"writeControl": { # Provides control over how write requests are executed. # Provides control over how write requests are executed. "requiredRevisionId": "A String", # The optional revision ID of the document the write request is applied to. If this is not the latest revision of the document, the request is not processed and returns a 400 bad request error. When a required revision ID is returned in a response, it indicates the revision ID of the document after the request was applied. "targetRevisionId": "A String", # The optional target revision ID of the document the write request is applied to. If collaborator changes have occurred after the document was read using the API, the changes produced by this write request are applied against the collaborator changes. This results in a new revision of the document that incorporates both the collaborator changes and the changes in the request, with the Docs server resolving conflicting changes. When using target revision ID, the API client can be thought of as another collaborator of the document. The target revision ID can only be used to write to recent versions of a document. If the target revision is too far behind the latest revision, the request is not processed and returns a 400 bad request error. The request should be tried again after retrieving the latest version of the document. Usually a revision ID remains valid for use as a target revision for several minutes after it's read, but for frequently edited documents this window might be shorter. - "writeMode": "A String", # How the request updates should be applied to the document. If unspecified, the request updates will be applied as normal edits. [Developer Preview](https://developers.google.com/workspace/preview). + "writeMode": "A String", # How the request updates should be applied to the document. If unspecified, the request updates will be applied as normal edits. }, } @@ -1436,14 +1436,14 @@Method Details
An object of the form: { # Response message from a BatchUpdateDocument request. - "commentUpdateState": "A String", # Whether comment updates were applied in the batch request. [Developer Preview](https://developers.google.com/workspace/preview). + "commentUpdateState": "A String", # Whether comment updates were applied in the batch request. "documentId": "A String", # The ID of the document to which the updates were applied to. "replies": [ # The reply of the updates. This maps 1:1 with the updates, although replies to some requests may be empty. { # A single response from an update. - "addCommentReply": { # Response message for adding a reply. [Developer Preview](https://developers.google.com/workspace/preview). # The result of adding a reply to a comment or suggestion. [Developer Preview](https://developers.google.com/workspace/preview). - "post": { # Represents a single post in a comment or suggestion thread. [Developer Preview](https://developers.google.com/workspace/preview). # The newly-inserted reply Post. + "addCommentReply": { # Response message for adding a reply. # The result of adding a reply to a comment or suggestion. + "post": { # Represents a single post in a comment or suggestion thread. # The newly-inserted reply Post. "assigneeEmail": "A String", # Optional. The email of the user who is being newly assigned to the thread as part of this post. Returns a 400 bad request error if: - The parent thread is a CommentThread whose headPost does not have an assignee. - The parent thread is a SuggestionThread. - commentAction is specified as `RESOLVE` or `REOPEN`. - `assigneeEmail` exceeds 2048 UTF-8 code units. - "author": { # Represents a user who authored a comment or suggestion post. [Developer Preview](https://developers.google.com/workspace/preview). # Output only. The user who created the post. + "author": { # Represents a user who authored a comment or suggestion post. # Output only. The user who created the post. "anonymous": True or False, # Whether the user is anonymous. "displayName": "A String", # The display name of the user. May be absent if the author is anonymous. "me": True or False, # Whether the user is the authenticated user making the request. @@ -1613,13 +1613,13 @@Method Details
"createNamedRange": { # The result of creating a named range. # The result of creating a named range. "namedRangeId": "A String", # The ID of the created named range. }, - "insertComment": { # Response message for inserting a comment. [Developer Preview](https://developers.google.com/workspace/preview). # The result of inserting a comment. [Developer Preview](https://developers.google.com/workspace/preview). - "commentThread": { # Represents a single comment thread. [Developer Preview](https://developers.google.com/workspace/preview). # The newly-inserted comment thread. + "insertComment": { # Response message for inserting a comment. # The result of inserting a comment. + "commentThread": { # Represents a single comment thread. # The newly-inserted comment thread. "anchorId": "A String", # The ID of the CommentAnchor in the document that this thread is tied to. Multiple comment threads may be anchored to the same CommentAnchor. "commentId": "A String", # The unique ID of the comment thread. - "headPost": { # Represents a single post in a comment or suggestion thread. [Developer Preview](https://developers.google.com/workspace/preview). # The first post in the thread. + "headPost": { # Represents a single post in a comment or suggestion thread. # The first post in the thread. "assigneeEmail": "A String", # Optional. The email of the user who is being newly assigned to the thread as part of this post. Returns a 400 bad request error if: - The parent thread is a CommentThread whose headPost does not have an assignee. - The parent thread is a SuggestionThread. - commentAction is specified as `RESOLVE` or `REOPEN`. - `assigneeEmail` exceeds 2048 UTF-8 code units. - "author": { # Represents a user who authored a comment or suggestion post. [Developer Preview](https://developers.google.com/workspace/preview). # Output only. The user who created the post. + "author": { # Represents a user who authored a comment or suggestion post. # Output only. The user who created the post. "anonymous": True or False, # Whether the user is anonymous. "displayName": "A String", # The display name of the user. May be absent if the author is anonymous. "me": True or False, # Whether the user is the authenticated user making the request. @@ -1639,9 +1639,9 @@Method Details
}, "plainTextQuote": "A String", # The quoted text from the document when the comment was created, formatted as plain-text. "replies": [ # Replies to the head post. - { # Represents a single post in a comment or suggestion thread. [Developer Preview](https://developers.google.com/workspace/preview). + { # Represents a single post in a comment or suggestion thread. "assigneeEmail": "A String", # Optional. The email of the user who is being newly assigned to the thread as part of this post. Returns a 400 bad request error if: - The parent thread is a CommentThread whose headPost does not have an assignee. - The parent thread is a SuggestionThread. - commentAction is specified as `RESOLVE` or `REOPEN`. - `assigneeEmail` exceeds 2048 UTF-8 code units. - "author": { # Represents a user who authored a comment or suggestion post. [Developer Preview](https://developers.google.com/workspace/preview). # Output only. The user who created the post. + "author": { # Represents a user who authored a comment or suggestion post. # Output only. The user who created the post. "anonymous": True or False, # Whether the user is anonymous. "displayName": "A String", # The display name of the user. May be absent if the author is anonymous. "me": True or False, # Whether the user is the authenticated user making the request. @@ -1815,8 +1815,8 @@Method Details
}, }, ], - "suggestionResponses": [ # The suggestions which were affected by each update. This maps 1:1 with the updates. [Developer Preview](https://developers.google.com/workspace/preview). - { # The suggestions which were affected by a given update. [Developer Preview](https://developers.google.com/workspace/preview). + "suggestionResponses": [ # The suggestions which were affected by each update. This maps 1:1 with the updates. + { # The suggestions which were affected by a given update. "acceptedSuggestionIds": [ # The IDs of suggestions which were accepted during the update. "A String", ], @@ -1837,7 +1837,7 @@Method Details
"writeControl": { # Provides control over how write requests are executed. # The updated write control after applying the request. "requiredRevisionId": "A String", # The optional revision ID of the document the write request is applied to. If this is not the latest revision of the document, the request is not processed and returns a 400 bad request error. When a required revision ID is returned in a response, it indicates the revision ID of the document after the request was applied. "targetRevisionId": "A String", # The optional target revision ID of the document the write request is applied to. If collaborator changes have occurred after the document was read using the API, the changes produced by this write request are applied against the collaborator changes. This results in a new revision of the document that incorporates both the collaborator changes and the changes in the request, with the Docs server resolving conflicting changes. When using target revision ID, the API client can be thought of as another collaborator of the document. The target revision ID can only be used to write to recent versions of a document. If the target revision is too far behind the latest revision, the request is not processed and returns a 400 bad request error. The request should be tried again after retrieving the latest version of the document. Usually a revision ID remains valid for use as a target revision for several minutes after it's read, but for frequently edited documents this window might be shorter. - "writeMode": "A String", # How the request updates should be applied to the document. If unspecified, the request updates will be applied as normal edits. [Developer Preview](https://developers.google.com/workspace/preview). + "writeMode": "A String", # How the request updates should be applied to the document. If unspecified, the request updates will be applied as normal edits. }, }
create(parent, body=None, volumeId=None, x__xgafv=None)
Creates a volume.
+
+ createFromSource(parent, body=None, x__xgafv=None)
Creates a volume in a volume pool from an existing data source (such as a Filestore backup).
Deletes a volume.
+
+ directCreate(parent, body=None, volumeId=None, x__xgafv=None)
Creates a volume directly on the CLH, bypassing CCFE. This is a temporary PoC that will be used for testing and removed.
+
+ directDelete(name, body=None, x__xgafv=None)
Deletes a volume directly on the CLH, bypassing CCFE. This is a temporary PoC that will be used for testing and removed.
Gets the details of a specific volume.
@@ -92,6 +101,12 @@
list_next(previous_request, previous_response)
Retrieves the next page of results.
+
+ patch(name, body=None, updateMask=None, x__xgafv=None)
Updates the settings of a specific volume.
+
+ unfreeze(name, body=None, x__xgafv=None)
Unfreezes a volume.
close()
@@ -110,6 +125,21 @@ createFromSource(parent, body=None, x__xgafv=None)
+ Creates a volume in a volume pool from an existing data source (such as a Filestore backup).
+
+Args:
+ parent: string, Required. The parent resource where this volume will be created. Format: projects/{project}/locations/{location}/volumePools/{volume_pool} (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request for creating a volume from an existing data source (such as a backup).
+ "backup": "A String", # The resource name of the backup to create from. Format: projects/{project}/locations/{location}/backups/{backup}
+ "ipAcls": [ # Optional. IP ACLs defining permissions/ACLs to apply to the newly created volume.
+ { # NFS export options specifications.
+ "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE.
+ "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions.
+ "A String",
+ ],
+ "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field.
+ "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS.
+ "A String",
+ ],
+ "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH.
+ },
+ ],
+ "quota": "A String", # Optional. The limit for how much data can be written to this volume, in MiB.
+ "volumeId": "A String", # Required. Unique identifier for the volume which also appears in the volumes resource name.
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # This resource represents a long-running operation that is the result of a network API call.
+ "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available.
+ "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation.
+ "code": 42, # The status code, which should be an enum value of google.rpc.Code.
+ "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use.
+ {
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ ],
+ "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.
+ },
+ "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`.
+ "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
}
directCreate(parent, body=None, volumeId=None, x__xgafv=None)
+ Creates a volume directly on the CLH, bypassing CCFE. This is a temporary PoC that will be used for testing and removed.
+
+Args:
+ parent: string, Required. The parent volume pool path, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}`. (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Volume representation of a Cloud Filestore volume.
+ "createTime": "A String", # Output only. The time when the volume was created.
+ "description": "A String", # Optional. A description of the volume with 2048 characters or less. Requests with longer descriptions will be rejected.
+ "ipAcls": [ # Optional. User-defined permissions protecting mount points.
+ { # NFS export options specifications.
+ "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE.
+ "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions.
+ "A String",
+ ],
+ "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field.
+ "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS.
+ "A String",
+ ],
+ "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH.
+ },
+ ],
+ "labels": { # Optional. Resource labels to represent user provided metadata.
+ "a_key": "A String",
+ },
+ "mountPoint": { # Mount details for a volume. # Output only. The mount point of the volume.
+ "ipAddress": "A String", # Output only. The IP address of the physical Filestore instance hosting the volume.
+ "mountName": "A String", # Output only. The mount name of the volume. Must be 63 characters or less and consist of uppercase or lowercase letters, numbers, and underscores.
+ },
+ "name": "A String", # Identifier. The resource name of the volume, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`.
+ "quotaMib": "A String", # Optional. The maximum size of the volume in mebibytes (MiB).
+ "state": "A String", # Optional. The state of the volume. This field is caller-settable via `UpdateVolume` to transition a volume from `HOT` to `COLD` (freezing the volume). In `CreateVolume`, this field is output-only and initialized to `HOT`.
+}
+
+ volumeId: string, Required. The ID to use for the volume. The ID must be unique within the specified volume pool.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Volume representation of a Cloud Filestore volume.
+ "createTime": "A String", # Output only. The time when the volume was created.
+ "description": "A String", # Optional. A description of the volume with 2048 characters or less. Requests with longer descriptions will be rejected.
+ "ipAcls": [ # Optional. User-defined permissions protecting mount points.
+ { # NFS export options specifications.
+ "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE.
+ "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions.
+ "A String",
+ ],
+ "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field.
+ "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS.
+ "A String",
+ ],
+ "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH.
+ },
+ ],
+ "labels": { # Optional. Resource labels to represent user provided metadata.
+ "a_key": "A String",
+ },
+ "mountPoint": { # Mount details for a volume. # Output only. The mount point of the volume.
+ "ipAddress": "A String", # Output only. The IP address of the physical Filestore instance hosting the volume.
+ "mountName": "A String", # Output only. The mount name of the volume. Must be 63 characters or less and consist of uppercase or lowercase letters, numbers, and underscores.
+ },
+ "name": "A String", # Identifier. The resource name of the volume, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`.
+ "quotaMib": "A String", # Optional. The maximum size of the volume in mebibytes (MiB).
+ "state": "A String", # Optional. The state of the volume. This field is caller-settable via `UpdateVolume` to transition a volume from `HOT` to `COLD` (freezing the volume). In `CreateVolume`, this field is output-only and initialized to `HOT`.
+}
+directDelete(name, body=None, x__xgafv=None)
+ Deletes a volume directly on the CLH, bypassing CCFE. This is a temporary PoC that will be used for testing and removed.
+
+Args:
+ name: string, Required. The volume resource name, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`. (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # DeleteVolumeRequest deletes a volume.
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); }
+}
+get(name, x__xgafv=None)
Gets the details of a specific volume. @@ -178,6 +389,21 @@Method Details
{ # Volume representation of a Cloud Filestore volume. "createTime": "A String", # Output only. The time when the volume was created. "description": "A String", # Optional. A description of the volume with 2048 characters or less. Requests with longer descriptions will be rejected. + "ipAcls": [ # Optional. User-defined permissions protecting mount points. + { # NFS export options specifications. + "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE. + "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings. + "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings. + "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions. + "A String", + ], + "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field. + "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS. + "A String", + ], + "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH. + }, + ], "labels": { # Optional. Resource labels to represent user provided metadata. "a_key": "A String", }, @@ -186,6 +412,8 @@Method Details
"mountName": "A String", # Output only. The mount name of the volume. Must be 63 characters or less and consist of uppercase or lowercase letters, numbers, and underscores. }, "name": "A String", # Identifier. The resource name of the volume, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`. + "quotaMib": "A String", # Optional. The maximum size of the volume in mebibytes (MiB). + "state": "A String", # Optional. The state of the volume. This field is caller-settable via `UpdateVolume` to transition a volume from `HOT` to `COLD` (freezing the volume). In `CreateVolume`, this field is output-only and initialized to `HOT`. }
patch(name, body=None, updateMask=None, x__xgafv=None)
+ Updates the settings of a specific volume.
+
+Args:
+ name: string, Identifier. The resource name of the volume, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`. (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Volume representation of a Cloud Filestore volume.
+ "createTime": "A String", # Output only. The time when the volume was created.
+ "description": "A String", # Optional. A description of the volume with 2048 characters or less. Requests with longer descriptions will be rejected.
+ "ipAcls": [ # Optional. User-defined permissions protecting mount points.
+ { # NFS export options specifications.
+ "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE.
+ "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions.
+ "A String",
+ ],
+ "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field.
+ "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS.
+ "A String",
+ ],
+ "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH.
+ },
+ ],
+ "labels": { # Optional. Resource labels to represent user provided metadata.
+ "a_key": "A String",
+ },
+ "mountPoint": { # Mount details for a volume. # Output only. The mount point of the volume.
+ "ipAddress": "A String", # Output only. The IP address of the physical Filestore instance hosting the volume.
+ "mountName": "A String", # Output only. The mount name of the volume. Must be 63 characters or less and consist of uppercase or lowercase letters, numbers, and underscores.
+ },
+ "name": "A String", # Identifier. The resource name of the volume, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`.
+ "quotaMib": "A String", # Optional. The maximum size of the volume in mebibytes (MiB).
+ "state": "A String", # Optional. The state of the volume. This field is caller-settable via `UpdateVolume` to transition a volume from `HOT` to `COLD` (freezing the volume). In `CreateVolume`, this field is output-only and initialized to `HOT`.
+}
+
+ updateMask: string, Required. Mask of fields to update. At least one path must be supplied in this field.
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # Volume representation of a Cloud Filestore volume.
+ "createTime": "A String", # Output only. The time when the volume was created.
+ "description": "A String", # Optional. A description of the volume with 2048 characters or less. Requests with longer descriptions will be rejected.
+ "ipAcls": [ # Optional. User-defined permissions protecting mount points.
+ { # NFS export options specifications.
+ "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE.
+ "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions.
+ "A String",
+ ],
+ "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field.
+ "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS.
+ "A String",
+ ],
+ "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH.
+ },
+ ],
+ "labels": { # Optional. Resource labels to represent user provided metadata.
+ "a_key": "A String",
+ },
+ "mountPoint": { # Mount details for a volume. # Output only. The mount point of the volume.
+ "ipAddress": "A String", # Output only. The IP address of the physical Filestore instance hosting the volume.
+ "mountName": "A String", # Output only. The mount name of the volume. Must be 63 characters or less and consist of uppercase or lowercase letters, numbers, and underscores.
+ },
+ "name": "A String", # Identifier. The resource name of the volume, in the format `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`.
+ "quotaMib": "A String", # Optional. The maximum size of the volume in mebibytes (MiB).
+ "state": "A String", # Optional. The state of the volume. This field is caller-settable via `UpdateVolume` to transition a volume from `HOT` to `COLD` (freezing the volume). In `CreateVolume`, this field is output-only and initialized to `HOT`.
+}
+unfreeze(name, body=None, x__xgafv=None)
+ Unfreezes a volume.
+
+Args:
+ name: string, Required. The name of the volume to unfreeze. Format: `projects/{project}/locations/{location}/volumePools/{volume_pool}/volumes/{volume}`. (required)
+ body: object, The request body.
+ The object takes the form of:
+
+{ # Request for unfreezing a volume.
+ "ipAcls": [ # Optional. IP ACLs defining permissions/ACLs to apply to the unfrozen volume
+ { # NFS export options specifications.
+ "accessMode": "A String", # Either READ_ONLY, for allowing only read requests on the exported directory, or READ_WRITE, for allowing both read and write requests. The default is READ_WRITE.
+ "anonGid": "A String", # An integer representing the anonymous group id with a default value of 65534. Anon_gid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "anonUid": "A String", # An integer representing the anonymous user id with a default value of 65534. Anon_uid may only be set with squash_mode of ROOT_SQUASH. An error will be returned if this field is specified for other squash_mode settings.
+ "ipRanges": [ # List of either an IPv4 addresses in the format `{octet1}.{octet2}.{octet3}.{octet4}` or CIDR ranges in the format `{octet1}.{octet2}.{octet3}.{octet4}/{mask size}` which may mount the file share. Overlapping IP ranges are not allowed, both within and across NfsExportOptions. An error will be returned. The limit is 64 IP ranges/addresses for each FileShareConfig among all NfsExportOptions.
+ "A String",
+ ],
+ "network": "A String", # Optional. The source VPC network for ip_ranges. Required for instances using Private Service Connect, optional otherwise. If provided, must be the same network specified in the `NetworkConfig.network` field.
+ "securityFlavors": [ # The security flavors allowed for mount operations. The default is AUTH_SYS.
+ "A String",
+ ],
+ "squashMode": "A String", # Either NO_ROOT_SQUASH, for allowing root access on the exported directory, or ROOT_SQUASH, for not allowing root access. The default is NO_ROOT_SQUASH.
+ },
+ ],
+ "quota": "A String", # Optional. The limit for how much data can be written to this volume, in MiB.
+}
+
+ x__xgafv: string, V1 error format.
+ Allowed values
+ 1 - v1 error format
+ 2 - v2 error format
+
+Returns:
+ An object of the form:
+
+ { # This resource represents a long-running operation that is the result of a network API call.
+ "done": True or False, # If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available.
+ "error": { # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). # The error result of the operation in case of failure or cancellation.
+ "code": 42, # The status code, which should be an enum value of google.rpc.Code.
+ "details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use.
+ {
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ ],
+ "message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.
+ },
+ "metadata": { # Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+ "name": "A String", # The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`.
+ "response": { # The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`.
+ "a_key": "", # Properties of the object. Contains field @type with type URL.
+ },
+}
+