Hailuo AI / MiniMax

opencollection: 1.0.0
info:
  name: Hailuo AI / MiniMax API
  version: '1.0'
request:
  auth:
    type: bearer
    token: '{{apiKey}}'
items:
- info:
    name: Video Generation
    type: folder
  items:
  - info:
      name: Create a video generation task.
      type: http
    http:
      method: POST
      url: https://api.minimax.io/v1/video_generation
      body:
        type: json
        data: "{\n  \"model\": \"MiniMax-Hailuo-2.3\",\n  \"prompt\": \"A cinematic drone shot flying over a neon-lit city\
          \ at night.\",\n  \"duration\": 6,\n  \"resolution\": \"1080P\",\n  \"prompt_optimizer\": true\n}"
    docs: Submits a Hailuo video generation request (text-to-video, image-to-video, first-last-frame, or subject-reference)
      and returns a task_id.
  - info:
      name: Query a video generation task.
      type: http
    http:
      method: GET
      url: https://api.minimax.io/v1/query/video_generation?task_id=:task_id
      params:
      - name: task_id
        value: ''
        type: query
        description: The task_id returned when the task was created.
    docs: Polls task status. Returns Queueing/Preparing/Processing, then Success (with file_id) or Fail. Poll about every
      10 seconds.
- info:
    name: Files
    type: folder
  items:
  - info:
      name: Retrieve a generated file.
      type: http
    http:
      method: GET
      url: https://api.minimax.io/v1/files/retrieve?file_id=:file_id
      params:
      - name: file_id
        value: ''
        type: query
        description: The file_id returned by a successful query response.
    docs: Returns a temporary download URL for the finished MP4 (video URLs valid for about 9 hours).
- info:
    name: Text to Speech
    type: folder
  items:
  - info:
      name: Synthesize speech (T2A v2).
      type: http
    http:
      method: POST
      url: https://api.minimax.io/v1/t2a_v2
      body:
        type: json
        data: "{\n  \"model\": \"speech-2.6-hd\",\n  \"text\": \"Welcome to the API Evangelist network.\",\n  \"voice_setting\"\
          : { \"voice_id\": \"male-qn-qingse\", \"speed\": 1.0 },\n  \"audio_setting\": { \"format\": \"mp3\", \"sample_rate\"\
          : 32000 }\n}"
    docs: Synchronous text-to-speech. A real-time WebSocket variant is documented at wss://api.minimax.io/ws/v1/t2a_v2.
- info:
    name: Chat Completions
    type: folder
  items:
  - info:
      name: Create a chat completion.
      type: http
    http:
      method: POST
      url: https://api.minimax.io/v1/chat/completions
      body:
        type: json
        data: "{\n  \"model\": \"MiniMax-M1\",\n  \"messages\": [ { \"role\": \"user\", \"content\": \"Hello!\" } ]\n}"
    docs: OpenAI-compatible chat completions served by MiniMax LLMs.
- info:
    name: Music Generation
    type: folder
  items:
  - info:
      name: Generate a vocal song (modeled).
      type: http
    http:
      method: POST
      url: https://api.minimax.io/v1/music_generation
      body:
        type: json
        data: "{\n  \"model\": \"music-1.5\",\n  \"prompt\": \"Upbeat synthwave with driving drums.\",\n  \"lyrics\": \"[verse]\
          \ Neon lights over the city tonight\"\n}"
    docs: Generates a vocal song from a style prompt plus lyrics. Endpoint payload is modeled; reconcile against the live
      reference.