MiniMax H3 API

Complete API reference for MiniMax H3 video generation models.

Model Variants

This API supports three model variants:

Quick Comparison

Supported Aspect Ratios

Text to Video and Reference to Video support:

21:9, 16:9, 4:3, 1:1, 3:4, 9:16

Resolution Options

MiniMax H3 Text to Video

Endpoint

POST /api/v1/video/task

Request Parameters

Required

Example Request

javascriptCopy
const response = await fetch('https://api.flaq.ai/api/v1/video/task', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model_name: 'minimax-h3-text-to-video',
    prompt: 'A golden retriever running through a sunlit meadow',
    duration: 8,
    resolution: '2k',
    aspect_ratio: '16:9'
  })
});

const { data } = await response.json();
const taskId = data.task_id;

MiniMax H3 Image to Video

Endpoint

POST /api/v1/video/task

Request Parameters

Required

Optional

Note: Image to Video does not support aspect_ratio.

Example Request

javascriptCopy
{
  model_name: 'minimax-h3-image-to-video',
  image_url: 'https://example.com/first-frame.jpg',
  image_end_url: 'https://example.com/last-frame.jpg',
  prompt: 'Camera slowly pulls back to reveal the full landscape',
  duration: 8,
  resolution: '2k'
}

MiniMax H3 Reference to Video

Generate a new video from optional image, video, and audio references.

Endpoint

POST /api/v1/video/task

Request Parameters

Required

Reference Media

@ (AT) Reference Placeholders

The @ (AT) reference feature links reference media to the prompt through numbered <<<...>>> placeholders. Numbering starts at 1 independently for each media type and follows the order of the corresponding array.

For example, if the first image is a character reference and the second image is a scene reference, use <<<image_1>>> for the character and <<<image_2>>> for the scene in prompt. Each placeholder must point to an item that exists in its corresponding array.

Example Request

javascriptCopy
{
  model_name: 'minimax-h3-reference-to-video',
  prompt: 'Use image one for the subject, video one for the movement, and audio one for the atmosphere',
  duration: 8,
  resolution: '2k',
  aspect_ratio: '16:9',
  images: ['https://example.com/subject-reference.jpg'],
  videos: ['https://example.com/motion-reference.mp4'],
  audios: ['https://example.com/atmosphere-reference.mp3']
}

@ (AT) Reference-to-Video Request Example

javascriptCopy
{
  model_name: 'minimax-h3-reference-to-video',
  prompt: 'Place the explorer from <<<image_1>>> in the environment from <<<image_2>>>, following the camera movement in <<<video_1>>> and speaking with the reference voice from <<<audio_1>>>',
  duration: 10,
  resolution: '2k',
  aspect_ratio: '16:9',
  images: [
    'https://example.com/explorer-reference.jpg',
    'https://example.com/environment-reference.jpg'
  ],
  videos: ['https://example.com/camera-movement-reference.mp4'],
  audios: ['https://example.com/voice-reference.mp3']
}

Response Format

Initial Response

All models return a task_id for polling:

jsonCopy
{
  "code": 0,
  "message": "success",
  "data": {
    "task_id": "{task_id}",
    "response_url": "https://api.flaq.ai/api/v1/video/{task_id}",
    "task_status": "submitted"
  }
}

Polling Response

Poll GET /api/v1/video/{taskId} to check status:

jsonCopy
{
  "code": 0,
  "message": "success",
  "data": {
    "task_id": "{task_id}",
    "response_url": "https://api.flaq.ai/api/v1/video/{task_id}",
    "task_status": "succeed",
    "task_status_msg": null,
    "task_result": {
      "videos": [
        {
          "url": "https://example.com/generated-video.mp4"
        }
      ]
    }
  }
}

Status Values

• submitted: Task accepted and waiting to be processed
• processing: Video generation in progress
• succeed: Video generation completed
• failed: Video generation failed

Best Practices

1. Describe Motion Clearly: Include specific movement descriptions in prompts
2. Resolution Selection: Use 768p for faster processing, 2k for higher quality output
3. Image Quality: For image-to-video, use high-quality input images for best results
4. Reference Ordering: Keep images, videos, and audios in intentional order and refer to them consistently in the prompt
5. @ (AT) References: Use <<<image_N>>>, <<<video_N>>>, and <<<audio_N>>> placeholders when you need explicit control over which reference asset drives each part of the scene