Microsoft Azure AI Foundry Chat Completions API

openapi: 3.1.0
info:
  title: Azure AI Foundry Model Inference REST Chat Completions API
  version: '2024-10-21'
  description: 'Data-plane REST API for running inference against models deployed in Microsoft

    Azure AI Foundry (formerly Azure AI Studio). The endpoint exposes OpenAI-compatible

    routes (`/chat/completions`, `/completions`, `/embeddings`) under the

    `/openai/deployments/{deployment-id}` or `/openai/v1` path on a per-resource host.


    Authentication uses either an API key (`api-key` header) or a Microsoft Entra ID

    OAuth 2.0 bearer token (scope `https://ai.azure.com/.default`).

    '
  contact:
    name: Microsoft Foundry
    url: https://learn.microsoft.com/en-us/azure/ai-foundry/
servers:
- url: https://{resource}.services.ai.azure.com/openai
  description: Foundry inference endpoint (per-resource)
  variables:
    resource:
      default: my-foundry
      description: Your Foundry resource name
- url: https://{resource}.openai.azure.com/openai
  description: Compatible Azure OpenAI endpoint exposed by the Foundry resource
  variables:
    resource:
      default: my-foundry
      description: Your Foundry resource name
tags:
- name: Chat Completions
  description: Chat-formatted text generation
paths:
  /deployments/{deployment-id}/chat/completions:
    post:
      summary: Create a chat completion
      operationId: createChatCompletion
      tags:
      - Chat Completions
      parameters:
      - in: path
        name: deployment-id
        required: true
        schema:
          type: string
        description: Foundry deployment name
      - in: query
        name: api-version
        required: true
        schema:
          type: string
          default: '2024-10-21'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ChatCompletionRequest'
      responses:
        '200':
          description: Chat completion response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ChatCompletionResponse'
      security:
      - apiKey: []
      - entra: []
  /v1/chat/completions:
    post:
      summary: Create a chat completion (OpenAI v1-compatible route)
      description: 'OpenAI-compatible route. The model deployment is selected by the `model`

        field in the request body rather than the URL path.

        '
      operationId: createChatCompletionV1
      tags:
      - Chat Completions
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ChatCompletionRequest'
      responses:
        '200':
          description: Chat completion response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ChatCompletionResponse'
      security:
      - entra: []
      - apiKey: []
components:
  schemas:
    ChatCompletionRequest:
      type: object
      required:
      - messages
      properties:
        model:
          type: string
          description: Deployment name (for v1 route)
        messages:
          type: array
          items:
            $ref: '#/components/schemas/ChatMessage'
        temperature:
          type: number
        top_p:
          type: number
        n:
          type: integer
        stream:
          type: boolean
        max_tokens:
          type: integer
        max_completion_tokens:
          type: integer
        stop:
          type:
          - string
          - array
        presence_penalty:
          type: number
        frequency_penalty:
          type: number
        user:
          type: string
    ChatCompletionResponse:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
        created:
          type: integer
        model:
          type: string
        choices:
          type: array
          items:
            type: object
            properties:
              index:
                type: integer
              message:
                $ref: '#/components/schemas/ChatMessage'
              finish_reason:
                type: string
        usage:
          type: object
          properties:
            prompt_tokens:
              type: integer
            completion_tokens:
              type: integer
            total_tokens:
              type: integer
    ChatMessage:
      type: object
      required:
      - role
      properties:
        role:
          type: string
          enum:
          - system
          - user
          - assistant
          - tool
          - developer
        content:
          type:
          - string
          - array
          - 'null'
        name:
          type: string
  securitySchemes:
    apiKey:
      type: apiKey
      in: header
      name: api-key
      description: Foundry/Azure OpenAI resource API key
    entra:
      type: oauth2
      description: Microsoft Entra ID OAuth 2.0 bearer token (scope https://ai.azure.com/.default)
      flows:
        clientCredentials:
          tokenUrl: https://login.microsoftonline.com/common/oauth2/v2.0/token
          scopes:
            https://ai.azure.com/.default: Default Foundry inference scope