osmAPI Chat Completions API

openapi: 3.1.0
info:
  title: osmAPI Chat Completions API
  description: >-
    OpenAI-compatible chat completions endpoint routed through osmAPI's unified
    AI gateway. Supports multi-turn conversations, function calling, streaming,
    structured output, web search, response healing, and smart routing across
    OpenAI, Anthropic, Google, and 14+ LLM providers.
  version: 1.0.0
  contact:
    name: osmAPI
    url: https://www.osmapi.com/
servers:
  - url: https://api.osmapi.com/v1
    description: Production
security:
  - bearerAuth: []
paths:
  /chat/completions:
    post:
      operationId: createChatCompletion
      summary: Create chat completion
      description: >-
        Sends a chat completion request to the specified model through osmAPI's
        smart routing system. Compatible with the OpenAI chat completions format.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ChatCompletionRequest'
      responses:
        '200':
          description: Successful completion response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ChatCompletionResponse'
        '429':
          description: Rate limit exceeded
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '500':
          description: Server error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
      tags:
        - Chat
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: Bearer token using osmAPI key (osmint_XXXXXXXXXXXXXXXX)
  schemas:
    ChatCompletionRequest:
      type: object
      required:
        - model
        - messages
      properties:
        model:
          type: string
          description: The AI model identifier or "auto" for smart routing
          examples:
            - gpt-4o
            - auto
            - openai/gpt-4o
        messages:
          type: array
          items:
            $ref: '#/components/schemas/Message'
          description: Conversation history with role-content pairs
        temperature:
          type: number
          nullable: true
          description: Controls response randomness (0-2)
        max_tokens:
          type: integer
          nullable: true
          description: Maximum number of tokens to generate
        top_p:
          type: number
          nullable: true
          description: Nucleus sampling parameter
        frequency_penalty:
          type: number
          nullable: true
          description: Reduces token repetition
        presence_penalty:
          type: number
          nullable: true
          description: Encourages new topics
        stream:
          type: boolean
          default: false
          description: Enable streaming responses
        response_format:
          type: object
          description: Specifies output structure (JSON schema support)
        tools:
          type: array
          items:
            type: object
          description: Available functions for model invocation
        tool_choice:
          oneOf:
            - type: string
            - type: object
          description: Tool selection strategy (required, none, auto)
        reasoning_effort:
          type: string
          nullable: true
          enum:
            - minimal
            - low
            - medium
            - high
          description: Reasoning intensity for supported models
        free_models_only:
          type: boolean
          description: Routes only to cost-free models
        no_reasoning:
          type: boolean
          description: Excludes reasoning models from routing
        web_search:
          type: boolean
          description: Enables real-time web search grounding
        plugins:
          type: array
          items:
            type: string
          description: Extended functionality plugins (e.g., response-healing)
    Message:
      type: object
      required:
        - role
        - content
      properties:
        role:
          type: string
          enum:
            - system
            - user
            - assistant
            - tool
            - function
        content:
          oneOf:
            - type: string
            - type: array
              items:
                type: object
        name:
          type: string
        tool_call_id:
          type: string
        tool_calls:
          type: array
          items:
            type: object
    ChatCompletionResponse:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
        created:
          type: integer
        model:
          type: string
        choices:
          type: array
          items:
            $ref: '#/components/schemas/Choice'
        usage:
          $ref: '#/components/schemas/Usage'
        metadata:
          $ref: '#/components/schemas/Metadata'
    Choice:
      type: object
      properties:
        index:
          type: integer
        message:
          type: object
          properties:
            role:
              type: string
            content:
              type: string
            reasoning:
              type: string
            tool_calls:
              type: array
              items:
                type: object
            images:
              type: array
              items:
                type: string
        finish_reason:
          type: string
    Usage:
      type: object
      properties:
        prompt_tokens:
          type: integer
        completion_tokens:
          type: integer
        total_tokens:
          type: integer
        reasoning_tokens:
          type: integer
        cost_usd_total:
          type: number
        cost_usd_input:
          type: number
        cost_usd_output:
          type: number
    Metadata:
      type: object
      properties:
        requested_model:
          type: string
        used_model:
          type: string
        used_provider:
          type: string
    ErrorResponse:
      type: object
      properties:
        error:
          type: object
          properties:
            message:
              type: string
            type:
              type: string
            code:
              type: string
tags:
  - name: Chat