Martian Chat Completions API

openapi: 3.0.1
info:
  title: Martian Gateway API
  description: >-
    OpenAPI specification for the Martian Gateway, an LLM model router. The
    Gateway exposes an OpenAI-compatible chat completions endpoint, an Anthropic
    Messages-compatible endpoint, and a models listing endpoint. Requests are
    routed dynamically across a catalog of provider models. Models are addressed
    with a provider/model-name string (for example, openai/gpt-4.1-nano).
  termsOfService: https://www.withmartian.com
  contact:
    name: Martian
    url: https://www.withmartian.com
  version: '1.0'
servers:
  - url: https://api.withmartian.com
    description: Martian Gateway production server
paths:
  /v1/chat/completions:
    post:
      operationId: createChatCompletion
      tags:
        - Chat Completions
      summary: Create a routed chat completion
      description: >-
        OpenAI-compatible chat completions endpoint. Supports most parameters of
        the OpenAI Chat Completions API. The model is specified as a
        provider/model-name string, and Martian routes the request to the best
        underlying model.
      security:
        - bearerAuth: []
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ChatCompletionRequest'
      responses:
        '200':
          description: A chat completion response, or a stream of Server-Sent Events when stream is true.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ChatCompletionResponse'
            text/event-stream:
              schema:
                type: string
                description: Server-Sent Events stream of partial chat completion chunks when stream is true.
        '401':
          description: Authentication failed (missing or invalid API key).
        '429':
          description: Rate limit or quota exceeded.
  /v1/messages:
    post:
      operationId: createMessage
      tags:
        - Messages
      summary: Create a routed message (Anthropic-compatible)
      description: >-
        Anthropic Messages-compatible endpoint. Supports most parameters of the
        Anthropic Messages API and routes the request through the Martian
        Gateway. The model is specified as a provider/model-name string.
      security:
        - bearerAuth: []
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/MessagesRequest'
      responses:
        '200':
          description: An Anthropic-compatible message response.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MessagesResponse'
        '401':
          description: Authentication failed (missing or invalid API key).
        '429':
          description: Rate limit or quota exceeded.
  /v1/models:
    get:
      operationId: listModels
      tags:
        - Models
      summary: List supported models
      description: >-
        Returns a JSON object containing all models currently supported by the
        Martian Gateway. The returned identifiers are provider/model-name
        strings usable as routing targets.
      security:
        - bearerAuth: []
      responses:
        '200':
          description: A list of models supported by the Gateway.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ModelList'
        '401':
          description: Authentication failed (missing or invalid API key).
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: Provide your Martian API key as a Bearer token in the Authorization header.
  schemas:
    ChatMessage:
      type: object
      required:
        - role
        - content
      properties:
        role:
          type: string
          enum:
            - system
            - user
            - assistant
            - tool
          description: The role of the author of this message.
        content:
          type: string
          description: The contents of the message.
        name:
          type: string
          description: An optional name for the participant.
    ChatCompletionRequest:
      type: object
      required:
        - model
        - messages
      properties:
        model:
          type: string
          description: Routing target in provider/model-name format (for example, openai/gpt-4.1-nano).
          example: openai/gpt-4.1-nano
        messages:
          type: array
          description: A list of messages comprising the conversation so far.
          items:
            $ref: '#/components/schemas/ChatMessage'
        max_tokens:
          type: integer
          description: The maximum number of tokens to generate in the completion.
        temperature:
          type: number
          description: Sampling temperature.
        top_p:
          type: number
          description: Nucleus sampling probability mass.
        stream:
          type: boolean
          description: If true, partial deltas are sent as Server-Sent Events.
          default: false
        stop:
          type: array
          items:
            type: string
          description: Up to a few sequences where the API will stop generating further tokens.
    ChatCompletionResponse:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
          example: chat.completion
        created:
          type: integer
        model:
          type: string
          description: The provider/model-name that the router selected to serve the request.
        choices:
          type: array
          items:
            type: object
            properties:
              index:
                type: integer
              message:
                $ref: '#/components/schemas/ChatMessage'
              finish_reason:
                type: string
        usage:
          $ref: '#/components/schemas/Usage'
    MessagesRequest:
      type: object
      required:
        - model
        - messages
        - max_tokens
      properties:
        model:
          type: string
          description: Routing target in provider/model-name format.
          example: anthropic/claude-sonnet-4
        messages:
          type: array
          description: Anthropic-style input messages.
          items:
            $ref: '#/components/schemas/ChatMessage'
        max_tokens:
          type: integer
          description: The maximum number of tokens to generate.
        system:
          type: string
          description: An optional system prompt.
        temperature:
          type: number
          description: Sampling temperature.
        stream:
          type: boolean
          default: false
    MessagesResponse:
      type: object
      properties:
        id:
          type: string
        type:
          type: string
          example: message
        role:
          type: string
          example: assistant
        model:
          type: string
          description: The provider/model-name that the router selected.
        content:
          type: array
          items:
            type: object
            properties:
              type:
                type: string
              text:
                type: string
        usage:
          $ref: '#/components/schemas/Usage'
    ModelList:
      type: object
      properties:
        object:
          type: string
          example: list
        data:
          type: array
          items:
            $ref: '#/components/schemas/Model'
    Model:
      type: object
      properties:
        id:
          type: string
          description: The provider/model-name identifier for the model.
          example: openai/gpt-4.1-nano
        object:
          type: string
          example: model
        owned_by:
          type: string
          description: The upstream provider that owns the model.
    Usage:
      type: object
      properties:
        prompt_tokens:
          type: integer
        completion_tokens:
          type: integer
        total_tokens:
          type: integer