Featherless Completions API

openapi: 3.0.1
info:
  title: Featherless AI API
  description: >-
    OpenAI-compatible REST API for the Featherless AI serverless inference
    platform. A single Bearer-authenticated interface serves thousands of
    open-weight Hugging Face models for chat completions, legacy text
    completions, embeddings, and model discovery. Pricing is a flat monthly
    subscription with unlimited tokens rather than per-token billing.
  termsOfService: https://featherless.ai/terms
  contact:
    name: Featherless AI Support
    url: https://featherless.ai
  version: '1.0'
servers:
  - url: https://api.featherless.ai/v1
    description: OpenAI-compatible base URL for Featherless AI inference.
security:
  - bearerAuth: []
tags:
  - name: Chat
    description: OpenAI-compatible chat completions.
  - name: Completions
    description: OpenAI-compatible legacy text completions.
  - name: Embeddings
    description: OpenAI-compatible text embeddings.
  - name: Models
    description: Model catalog discovery.
paths:
  /chat/completions:
    post:
      operationId: createChatCompletion
      tags:
        - Chat
      summary: Create a chat completion
      description: >-
        Generates a model response for the given chat conversation. OpenAI
        compatible. Set `stream: true` to receive the response as
        Server-Sent Events. Vision is supported by passing image_url content
        parts in messages, and tool calling by supplying `tools`.
      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:
                type: string
                description: >-
                  When `stream` is true, a sequence of `data:` SSE events each
                  carrying a `chat.completion.chunk` object, terminated by
                  `data: [DONE]`.
        '401':
          $ref: '#/components/responses/Unauthorized'
        '429':
          $ref: '#/components/responses/TooManyRequests'
  /completions:
    post:
      operationId: createCompletion
      tags:
        - Completions
      summary: Create a text completion
      description: >-
        Generates a text completion for the provided prompt. OpenAI
        compatible. The `prompt` may be a string or an array of strings; an
        array is treated as a parallel inference call. Supports streaming via
        `stream: true`.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateCompletionRequest'
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CreateCompletionResponse'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '429':
          $ref: '#/components/responses/TooManyRequests'
  /embeddings:
    post:
      operationId: createEmbedding
      tags:
        - Embeddings
      summary: Create embeddings
      description: >-
        Returns vector embeddings for the supplied input using an open
        embedding model (for example Qwen/Qwen3-Embedding-8B). OpenAI
        compatible.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateEmbeddingRequest'
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CreateEmbeddingResponse'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '429':
          $ref: '#/components/responses/TooManyRequests'
  /models:
    get:
      operationId: listModels
      tags:
        - Models
      summary: List models
      description: >-
        Lists the open-weight models available for serverless inference on
        Featherless AI. OpenAI compatible.
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ListModelsResponse'
        '401':
          $ref: '#/components/responses/Unauthorized'
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      bearerFormat: Featherless API key
      description: >-
        Set `Authorization: Bearer <FEATHERLESS_API_KEY>`. Keys are created
        in the Featherless dashboard.
  responses:
    Unauthorized:
      description: Missing or invalid API key.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
    TooManyRequests:
      description: >-
        Concurrency limit for the account's plan exceeded. Featherless meters
        concurrent connections rather than tokens.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
  schemas:
    CreateChatCompletionRequest:
      type: object
      required:
        - model
        - messages
      properties:
        model:
          type: string
          description: Hugging Face model identifier, e.g. `Qwen/Qwen2.5-7B-Instruct`.
        messages:
          type: array
          description: A list of messages comprising the conversation so far.
          items:
            $ref: '#/components/schemas/ChatMessage'
        temperature:
          type: number
          nullable: true
          description: Sampling temperature.
        top_p:
          type: number
          nullable: true
          description: Nucleus sampling probability mass.
        top_k:
          type: integer
          nullable: true
          description: Limits sampling to the top-k tokens.
        min_p:
          type: number
          nullable: true
          description: Minimum token probability relative to the most likely token.
        presence_penalty:
          type: number
          nullable: true
        frequency_penalty:
          type: number
          nullable: true
        repetition_penalty:
          type: number
          nullable: true
        max_tokens:
          type: integer
          nullable: true
          description: Maximum number of tokens to generate.
        min_tokens:
          type: integer
          nullable: true
        stop:
          oneOf:
            - type: string
            - type: array
              items:
                type: string
          nullable: true
        seed:
          type: integer
          nullable: true
          description: Random seed; not guaranteed reproducible across servers.
        tools:
          type: array
          description: A list of tools (functions) the model may call.
          items:
            type: object
        stream:
          type: boolean
          default: false
          description: If true, partial deltas are streamed as Server-Sent Events.
    ChatMessage:
      type: object
      required:
        - role
        - content
      properties:
        role:
          type: string
          enum:
            - system
            - user
            - assistant
            - tool
        content:
          description: >-
            String content, or an array of content parts (text and image_url)
            for vision-capable models.
          oneOf:
            - type: string
            - type: array
              items:
                type: object
    CreateChatCompletionResponse:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
          enum:
            - chat.completion
        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
                nullable: true
        usage:
          $ref: '#/components/schemas/Usage'
    CreateCompletionRequest:
      type: object
      required:
        - model
        - prompt
      properties:
        model:
          type: string
          description: Hugging Face model identifier.
        prompt:
          description: >-
            Prompt to complete. A string, or an array of strings treated as a
            parallel inference call.
          oneOf:
            - type: string
            - type: array
              items:
                type: string
        temperature:
          type: number
          nullable: true
        top_p:
          type: number
          nullable: true
        top_k:
          type: integer
          nullable: true
        min_p:
          type: number
          nullable: true
        repetition_penalty:
          type: number
          nullable: true
        max_tokens:
          type: integer
          nullable: true
        stop:
          oneOf:
            - type: string
            - type: array
              items:
                type: string
          nullable: true
        stream:
          type: boolean
          default: false
    CreateCompletionResponse:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
          enum:
            - text_completion
        created:
          type: integer
        model:
          type: string
        choices:
          type: array
          items:
            type: object
            properties:
              index:
                type: integer
              text:
                type: string
              finish_reason:
                type: string
                nullable: true
        usage:
          $ref: '#/components/schemas/Usage'
    CreateEmbeddingRequest:
      type: object
      required:
        - model
        - input
      properties:
        model:
          type: string
          description: Embedding model identifier, e.g. `Qwen/Qwen3-Embedding-8B`.
        input:
          description: A string or array of strings to embed.
          oneOf:
            - type: string
            - type: array
              items:
                type: string
        encoding_format:
          type: string
          enum:
            - float
            - base64
          default: float
        dimensions:
          type: integer
          nullable: true
          description: Output dimensionality, for models that support it.
    CreateEmbeddingResponse:
      type: object
      properties:
        object:
          type: string
          enum:
            - list
        model:
          type: string
        data:
          type: array
          items:
            type: object
            properties:
              object:
                type: string
                enum:
                  - embedding
              index:
                type: integer
              embedding:
                type: array
                items:
                  type: number
        usage:
          type: object
          properties:
            prompt_tokens:
              type: integer
            total_tokens:
              type: integer
    ListModelsResponse:
      type: object
      properties:
        object:
          type: string
          enum:
            - list
        data:
          type: array
          items:
            $ref: '#/components/schemas/Model'
    Model:
      type: object
      properties:
        id:
          type: string
          description: Model identifier.
        object:
          type: string
          enum:
            - model
        created:
          type: integer
        owned_by:
          type: string
    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
              nullable: true