get_ai_model_details

ArtificialIntelligenceApi.get_ai_model_details(model_id, role_id, action=None, invocation_id=None, _request_timeout=None, _request_auth=None, _content_type=None, _headers=None, _host_index=0)

Get model details or invocation status

Two distinct reads on the same path, selected by the action query parameter. The two modes return different fields; both are described by AIModelDetails. action=get_details (the default when action is omitted) returns the model’s metadata: identity, status, parameters, modalities, supported inference types, adaptive-thinking capability, the Amorphic components the model is allow-listed for, and its cost tags. AdaptiveThinking is always present as a boolean. Effort is always present (an empty array for models without adaptive thinking). CostTags is always present and is an empty array when cost management is disabled for the environment. action=invoke returns the state of a single asynchronous invocation started by POST /ai/models/{model_id} and requires invocation_id. It returns UserInput, InvocationStatus, and InvocationTime, plus ModelResponse once the invocation has completed. No model metadata is returned in this mode. Poll until InvocationStatus is no longer IN_PROGRESS. Invocation records expire 30 days after creation; after that, lookup fails as an invalid invocation id. Any other action value returns GE-1010 as a 400.

Parameters:

  • model_id (str) – Bedrock model identifier as stored by Amorphic, for example anthropic.claude-sonnet-4-6. With action=get_details, an unknown model_id returns IPV-1002 as a 500 (not a 404). With action=invoke, model_id is part of the invocation record key and must match the model the invocation was started against. (required)

  • role_id (str) – User role id (required)

  • action (str) – Read mode. get_details (the default) returns model metadata. invoke returns the status and result of a prior asynchronous invocation and requires invocation_id. Any other value returns GE-1010 as a 400.

  • invocation_id (str) – Required when action=invoke; the InvocationId returned by POST /ai/models/{model_id}. Omitting it returns IPV-1001 and an unknown or expired id returns IPV-1002, both as a 400. Ignored when action=get_details.

  • _request_timeout (int, tuple(int, int), optional) – timeout setting for this request. If one number provided, it will be total request timeout. It can also be a pair (tuple) of (connection, read) timeouts.

  • _request_auth (dict, optional) – set to override the auth_settings for an a single request; this effectively ignores the authentication in the spec for a single request.

  • _content_type (str, Optional) – force content-type for the request.

  • _headers (dict, optional) – set to override the headers for a single request; this effectively ignores the headers in the spec for a single request.

  • _host_index (int, optional) – set to override the host_index for a single request; this effectively ignores the host_index in the spec for a single request.

Return type:

AIModelDetails

Returns:

Returns the result object.

Request and Response Examples

HTTP: GET /ai/models/{model_id}

Success (HTTP 200)

Model metadata when action=get_details, or invocation state when action=invoke. ModelResponse is present only once the invocation has completed successfully.

action=get_details on an enabled model

{
  "ModelId": "anthropic.claude-sonnet-4-6",
  "ModelName": "Claude Sonnet 4.6",
  "ModelVersion": "1.0",
  "ModelTokenLimit": "200000",
  "ModelArn": "arn:aws:bedrock:us-east-1::foundation-model/anthropic.claude-sonnet-4-6",
  "ModelStatus": "Enabled",
  "ProviderName": "Anthropic",
  "ModelStatusMessage": "Model is enabled and ready to use.",
  "ModelType": "BASE",
  "ModelDescription": "General purpose reasoning model.",
  "ModelTraits": "Reasoning, Long context",
  "ModelParameters": {
    "temperature": 0.5,
    "top_p": 0.9,
    "max_tokens": 4096
  },
  "AdaptiveThinking": true,
  "Effort": [
    "low",
    "medium",
    "high"
  ],
  "InputModalities": [
    "TEXT",
    "IMAGE"
  ],
  "OutputModalities": [
    "TEXT"
  ],
  "InferenceTypesSupported": [
    "ON_DEMAND",
    "INFERENCE_PROFILE"
  ],
  "DefaultInferenceType": "INFERENCE_PROFILE",
  "InferenceProfileDetails": {
    "DefaultInferenceProfile": "arn:aws:bedrock:us-east-1:000000000000:inference-profile/us.anthropic.claude-sonnet-4-6",
    "InferenceProfiles": [
      "arn:aws:bedrock:us-east-1:000000000000:inference-profile/us.anthropic.claude-sonnet-4-6"
    ],
    "ModelArns": [
      "arn:aws:bedrock:us-east-1::foundation-model/anthropic.claude-sonnet-4-6"
    ]
  },
  "Components": [
    {
      "ComponentName": "agents",
      "DisplayName": "Agents"
    },
    {
      "ComponentName": "nl2sql",
      "DisplayName": "SQL AI"
    }
  ],
  "CostTags": [
    {
      "TagKey": "BusinessUnit",
      "TagValue": "analytics"
    }
  ],
  "LastModifiedTime": "2026-08-04 11:20:31",
  "LastModifiedBy": "ai.platform.admin"
}

action=invoke, generation still running

{
  "UserInput": "Summarize the quarterly revenue trend.",
  "InvocationStatus": "IN_PROGRESS",
  "InvocationTime": "2026-08-11 09:14:02"
}

action=invoke, generation finished

{
  "UserInput": "Summarize the quarterly revenue trend.",
  "InvocationStatus": "COMPLETED",
  "InvocationTime": "2026-08-11 09:14:02",
  "ModelResponse": "Revenue grew steadily across the quarter, with the strongest month at the end of the period."
}

Errors

Documented error codes: IPV-1001, IPV-1002, GE-1010, AUTH-1014, DB-1001, GE-1008.

HTTP 400

Input validation failure. Body is {“Message”: “<CODE> - <text>”}. Codes: IPV-1001 (invocation_id missing when action=invoke), IPV-1002 (invocation_id unknown, expired, or belonging to a different model), GE-1010 (action value is not get_details or invoke). AUTH-1014 from the role check is also returned here, not as a 403.

invocationIdMissing

{
  "Message": "IPV-1001 - invocation_id must be provided as a query parameter for model invocation GET request"
}

invocationIdInvalid

{
  "Message": "IPV-1002 - invocation ID provided is invalid"
}

HTTP 500

Backend failure. Body is {“Message”: “<CODE> - <text>”}. Codes: IPV-1002 (unknown model_id with action=get_details — returned as 500, not 404), DB-1001 (invocation lookup failed), GE-1008 (catch-all).

Unknown model_id with action=get_details

{
  "Message": "IPV-1002 - Invalid Model - anthropic.claude-does-not-exist, resource not found."
}

backendFailure

{
  "Message": "GE-1008 - Could not complete the request. Please try again."
}