openapi: 3.0.1
info:
  title: glhf Chat API
  description: OpenAI-compatible REST API for glhf (glhf.chat), a platform that runs almost any open-source large language model on demand. Models are selected by passing a Hugging Face repository identifier as hf:org/model. Only the endpoints documented for the glhf OpenAI-compatible surface are described here.
  termsOfService: https://glhf.chat
  contact:
    name: glhf
    url: https://glhf.chat
  version: '1.0'
servers:
- url: https://glhf.chat/api/openai/v1
  description: glhf OpenAI-compatible base URL.
security:
- bearerAuth: []
tags:
- name: Chat
paths:
  /chat/completions:
    post:
      operationId: createChatCompletion
      tags:
      - Chat
      summary: Create a chat completion.
      description: 'Creates a model response for the given chat conversation. The model field must be a Hugging Face repository in the form hf:org/model (for example hf:meta-llama/Llama-3.3-70B-Instruct). When stream is true, partial message deltas are returned as Server-Sent Events terminated by a data: [DONE] message.'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateChatCompletionRequest'
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CreateChatCompletionResponse'
            text/event-stream:
              schema:
                $ref: '#/components/schemas/ChatCompletionChunk'
        '400':
          description: Bad request, for example a model identifier without a valid prefix.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '401':
          description: Unauthorized, missing or invalid API key.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
components:
  schemas:
    ChatCompletionChunk:
      type: object
      description: A streamed chunk of a chat completion delivered as a Server-Sent Event when stream is true.
      properties:
        id:
          type: string
        object:
          type: string
          example: chat.completion.chunk
        created:
          type: integer
        model:
          type: string
        choices:
          type: array
          items:
            type: object
            properties:
              index:
                type: integer
              delta:
                $ref: '#/components/schemas/ChatMessage'
              finish_reason:
                type: string
                nullable: true
    CreateChatCompletionResponse:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
          example: chat.completion
        created:
          type: integer
        model:
          type: string
        choices:
          type: array
          items:
            $ref: '#/components/schemas/ChatCompletionChoice'
        usage:
          $ref: '#/components/schemas/Usage'
    Usage:
      type: object
      properties:
        prompt_tokens:
          type: integer
        completion_tokens:
          type: integer
        total_tokens:
          type: integer
    Error:
      type: object
      properties:
        error:
          type: object
          properties:
            message:
              type: string
            type:
              type: string
            code:
              type: string
    ChatMessage:
      type: object
      required:
      - role
      - content
      properties:
        role:
          type: string
          enum:
          - system
          - user
          - assistant
        content:
          type: string
    CreateChatCompletionRequest:
      type: object
      required:
      - model
      - messages
      properties:
        model:
          type: string
          description: Hugging Face repository identifier in the form hf:org/model.
          example: hf:meta-llama/Llama-3.3-70B-Instruct
        messages:
          type: array
          description: A list of messages comprising the conversation so far.
          items:
            $ref: '#/components/schemas/ChatMessage'
        stream:
          type: boolean
          description: If true, partial message deltas are sent as Server-Sent Events.
          default: false
        temperature:
          type: number
          format: float
          description: Sampling temperature between 0 and 2.
        top_p:
          type: number
          format: float
          description: Nucleus sampling probability mass.
        max_tokens:
          type: integer
          description: Maximum number of tokens to generate in the completion.
        stop:
          oneOf:
          - type: string
          - type: array
            items:
              type: string
          description: Up to a few sequences where generation will stop.
        n:
          type: integer
          description: Number of completions to generate for each prompt.
        presence_penalty:
          type: number
          format: float
        frequency_penalty:
          type: number
          format: float
    ChatCompletionChoice:
      type: object
      properties:
        index:
          type: integer
        message:
          $ref: '#/components/schemas/ChatMessage'
        finish_reason:
          type: string
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: glhf API key passed as a Bearer token in the Authorization header. Keys are created at https://glhf.chat/users/settings/api.