AI Integrations

Create an AI integration

Changed on

Create a new AI integration for an external LLM provider.

Payload Requirements

  • name and provider are required.
  • The integration name must be unique within the account.
  • provider must be one of: OPEN_AI, AZURE_OPEN_AI, AWS_BEDROCK, VERTEX_AI, ANTHROPIC, NVIDIA_NIM, GEMINI, CUSTOM, LITELLM, FIREWORKS, TOGETHER_AI.
  • If scopings is omitted, the integration defaults to account-wide visibility.
  • enable_default_models defaults to false if not provided.
  • function_calling_enabled defaults to true if not provided.
  • auth_type defaults to DEFAULT if not provided.
  • For AWS_BEDROCK provider, provider_metadata must include role_arn.
  • For VERTEX_AI provider, provider_metadata must include project_id, location, and project_access_label.

Valid example

{
  "name": "Production OpenAI",
  "provider": "OPEN_AI",
  "api_key": "sk-abc123...",
  "model_names": ["gpt-4", "gpt-4o"],
  "enable_default_models": true
}

Invalid example (missing required provider)

{
  "name": "My Integration"
}

<Warning>This endpoint is in alpha, read more here.</Warning>

post/v2/ai-integrations

Request

  • Base URL
    https://api.arize.com
  • URL
    https://api.arize.com/v2/ai-integrations
  • Auth
    HTTP bearer

Request body

namestring required

Integration name

provider'OPEN_AI' | 'AZURE_OPEN_AI' | 'AWS_BEDROCK' | 'VERTEX_AI' | 'ANTHROPIC' | 'CUSTOM' | 'NVIDIA_NIM' | 'GEMINI' | 'LITELLM' | 'FIREWORKS' | 'TOGETHER_AI' required

The AI provider for this integration

api_keystring

API key for the provider (write-only, never returned)

base_urlstring

Custom base URL for the provider

model_namesstring[]

Supported model names

headersobject

Custom headers to include in requests. The serialized header map must not exceed 8,175 bytes.

enable_default_modelsboolean

Enable provider's default model list (default false)

function_calling_enabledboolean

Enable function/tool calling (default true)

auth_type'DEFAULT' | 'PROXY_WITH_HEADERS' | 'BEARER_TOKEN' | 'OAUTH2_CLIENT_CREDENTIALS'

The authentication method for this integration. OAUTH2_CLIENT_CREDENTIALS is returned for integrations configured with OAuth 2.0 client credentials in the Arize UI. Setting it through this API is not yet supported.

Response

An AI integration object

idstring required

The integration ID

namestring required

The integration name

provider'OPEN_AI' | 'AZURE_OPEN_AI' | 'AWS_BEDROCK' | 'VERTEX_AI' | 'ANTHROPIC' | 'CUSTOM' | 'NVIDIA_NIM' | 'GEMINI' | 'LITELLM' | 'FIREWORKS' | 'TOGETHER_AI' required

The AI provider for this integration

has_api_keyboolean required

Whether an API key is configured (the key itself is never returned)

base_urlstring nullable

Custom base URL for the provider

model_namesstring[] nullable

Supported model names

headersobject nullable

Custom headers included in requests

enable_default_modelsboolean required

Whether the provider's default model list is enabled

function_calling_enabledboolean required

Whether function/tool calling is enabled

auth_type'DEFAULT' | 'PROXY_WITH_HEADERS' | 'BEARER_TOKEN' | 'OAUTH2_CLIENT_CREDENTIALS' required

The authentication method for this integration. OAUTH2_CLIENT_CREDENTIALS is returned for integrations configured with OAuth 2.0 client credentials in the Arize UI. Setting it through this API is not yet supported.

created_atstring date-time required

When the integration was created

updated_atstring date-time required

When the integration was last updated

created_by_user_idstring required

The user ID of the user who created the integration

Changes

  • 00294ff0845aRevision changes
    • warning

      added the new TOGETHER_AI enum value to the response property for the response status

    • info

      added the new TOGETHER_AI enum value to the request property

  • 15b98c7ec8ccRevision changes
    • warning

      added the new FIREWORKS enum value to the response property for the response status

    • info

      added the new FIREWORKS enum value to the request property