# MiniMax Hailuo-03 (H3) Reference to Video

> MiniMax Hailuo-03 (H3) reference-to-video via Fal.AI. Generates video from multimodal references — up to 5 subject/style images, 3 motion video clips, and 3 audio clips, each cited in the prompt by order (Image 1, Video 1, Audio 1) — keeping subjects consistent while following the referenced motion and audio. 5-15 second duration, 480P/768P native or 2K/4K upscaled output.

- **Provider**: fal
- **Model ID**: minimax/h3/reference-to-video
- **Category**: video_generation
- **Credits**: 1510 per request
- **Pricing Type**: input_based

## API Endpoint

Base URL: https://api.core.today/v1

### Create Prediction
POST /predictions

### Get Status
GET /predictions/{job_id}

### Cancel
DELETE /predictions/{job_id}

## Authentication

Header: `X-API-Key: YOUR_API_KEY`

## Input Parameters

- `duration` (integer, optional): The duration of the video in seconds. (Default: `5`; Range: min: 5, max: 15)
- `enable_safety_checker` (boolean, optional): If set to true, the safety checker will be enabled. (Default: `True`)
- `aspect_ratio` (string, optional): The aspect ratio of the generated video. (Default: `adaptive`; Options: `adaptive`, `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16`)
- `prompt_expansion_mode` (string, optional): How much effort to spend rewriting the prompt before generation. 'fast' returns in about a second. 'balanced' picks per request. 'quality' spends up to ~30s on a richer prompt. (Default: `balanced`; Options: `fast`, `balanced`, `quality`)
- `seed` (integer, optional): Random seed. A random seed is selected when omitted.
- `reference_audio_urls` (array, optional): URLs of reference audio clips (2-15 seconds each, combined duration at most 15 seconds), referenced in the prompt as Audio 1, Audio 2, and so on. Up to 3 clips. Audio cannot be the only reference input; provide at least one reference image or video with it.
- `prompt` (string, **required**): Text prompt for video generation. Refer to reference assets by their modality and order in the reference lists: Image 1, Image 2, Video 1, Audio 1, and so on.
- `resolution` (string, optional): The resolution of the generated video. 480P and 768P are native generation modes; 2K and 4K upscale a 768P base result. (Default: `2K`; Options: `480P`, `768P`, `2K`, `4K`)
- `reference_image_urls` (array, optional): URLs of subject/style reference images, referenced in the prompt as Image 1, Image 2, and so on. Up to 5 images (this gateway caps at the surcharge-free tier). Reference images, videos, and audio clips must add up to at most 12 files.
- `reference_video_urls` (array, optional): URLs of motion/reference video clips (2-15 seconds each, combined duration at most 15 seconds), referenced in the prompt as Video 1, Video 2, and so on. Up to 3 clips.

## Example Request

```json
{
  "model": "minimax/h3/reference-to-video",
  "input": {
    "duration": 5,
    "aspect_ratio": "adaptive",
    "prompt": "Image 1 is the female protagonist. Image 2 is her small dog. Keep the woman and dog consistent with their respective reference images while they walk together through a sunlit garden.",
    "resolution": "2K",
    "reference_image_urls": [
      "https://storage.googleapis.com/falserverless/example_inputs/hailuo23/pro_i2v_in.jpg"
    ]
  }
}
```

## Response Format

```json
{
  "job_id": "abc123",
  "status": "pending",
  "provider": "replicate",
  "model": "black-forest-labs/flux-schnell",
  "created_at": "2026-01-01T00:00:00Z",
  "result": null,
  "error": null
}
```

Status values: `pending`, `processing`, `completed`, `failed`, `cancelled`

## Usage Flow

1. POST /predictions with model and input → receive job_id
2. Poll GET /predictions/{job_id} until status is `completed` or `failed`
3. Result contains output URL(s) or data

## Output Type

json

## Tags

video-generation, reference-to-video, multimodal-reference, character-consistency, lip-sync

## Documentation

https://fal.ai/models/minimax/h3/reference-to-video