# MiniMax Hailuo-03 (H3) Image to Video - Core.Today AI API

> MiniMax Hailuo-03 (H3) image-to-video via Fal.AI. 2K video generation from a first-frame image, 5-15 second duration, native audio, with optional first-to-last keyframe control via end_image_url.

- **Provider**: MiniMax
- **Model ID**: minimax/h3/image-to-video
- **Category**: Video Generation
- **Credits**: 3020 per second, tiered by resolution
- **Speed**: Medium
- **Quality**: Ultra

## Features
- Generates video directly from a single first-frame image
- Up to 2K resolution output
- 5-15 second adjustable duration
- Native audio generated alongside the video
- Optional first-to-last keyframe control via end_image_url
- Output aspect ratio automatically follows the input image

## Use Cases
- Animating a static product or portrait photo into a short video clip
- Turning a storyboard frame into a cinematic camera-move sequence
- First-to-last keyframe transitions between two designed shots
- Social media teaser clips generated from a single hero image

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **prompt**: string - Text prompt for video generation
- **image_url**: string - URL of the image to use as the first frame. The output aspect ratio follows this image.

### Optional
- **duration**: integer (default: 5) - The duration of the video in seconds.
- **resolution**: string (default: 2K) - The resolution of the generated video.
  Options: 768P, 2K
- **end_image_url**: string - Optional URL of the image to use as the last frame, for first-to-last keyframe generation.

## Examples

### Slow pull-back reveal from a landscape photo
Animate a still landscape image into a 5-second 2K clip where the camera pulls back to reveal the full scene.
```json
{
  "model": "minimax/h3/image-to-video",
  "input": {
    "prompt": "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain.",
    "image_url": "https://storage.googleapis.com/falserverless/example_inputs/hailuo23/pro_i2v_in.jpg",
    "duration": 5,
    "resolution": "2K"
  }
}
```

### First-to-last keyframe transition
Provide both a starting image and an ending image so the model generates a longer transition video between the two.
```json
{
  "model": "minimax/h3/image-to-video",
  "input": {
    "prompt": "A smooth cinematic transition as the scene transforms from day to night.",
    "image_url": "https://example.com/start-frame.jpg",
    "end_image_url": "https://example.com/end-frame.jpg",
    "duration": 10,
    "resolution": "2K"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Use end_image_url when you need a controlled transition between two specific shots instead of an open-ended camera move.
- The output aspect ratio always follows image_url, so crop your source image to the target ratio before uploading.
- Longer durations (closer to 15 seconds) work best with prompts that describe a clear beginning-to-end camera or scene progression.
- Drop to 768P resolution for quick drafts or previews before committing to a full 2K render.

## Documentation
https://fal.ai/models/minimax/h3/image-to-video