# MiniMax H3 Image to Video LoRA

> Images into video with synchronized audio using MiniMax H3; your image becomes the first frame, prompt optional, with trained LoRA support for subject consistency.


## Overview

- **Endpoint**: `https://fal.run/minimax/h3/image-to-video/lora`
- **Model ID**: `minimax/h3/image-to-video/lora`
- **Category**: image-to-video
- **Kind**: inference
**Tags**: utility, editing



## Pricing

Video costs **$0.0625** per second at **480p**, **$0.075** per second at **768p**, **$0.1625** per second at **2K** and **$0.20** per second at **4K**.

For more details, see [fal.ai pricing](https://fal.ai/pricing).

## API Information

This model can be used via our HTTP API or more conveniently via our client libraries.
See the input and output schema below, as well as the usage examples.


### Input Schema

The API accepts the following input parameters:


- **`prompt`** (`string`, _required_):
  Text prompt for video generation
  - Examples: "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain."

- **`duration`** (`integer`, _optional_):
  The duration of the video in seconds. Default value: `5`
  - Default: `5`
  - Range: `5` to `15`

- **`resolution`** (`ResolutionEnum`, _optional_):
  The resolution of the generated video. 480P and 768P are native generation modes; 2K and 4K upscale a 768P base result. Default value: `"2K"`
  - Default: `"2K"`
  - Options: `"480P"`, `"768P"`, `"2K"`, `"4K"`

- **`seed`** (`integer`, _optional_):
  Random seed. A random seed is selected when omitted.

- **`enable_safety_checker`** (`boolean`, _optional_):
  If set to true, the safety checker will be enabled. Default value: `true`
  - Default: `true`

- **`sync_mode`** (`boolean`, _optional_):
  Return the generated video as base64 instead of a CDN URL.
  - Default: `false`

- **`prompt_expansion_mode`** (`string`, _optional_):
  How much effort to spend rewriting the prompt before generation. 'disabled' skips prompt expansion. 'fast' returns in about a second. 'balanced' picks per request. 'quality' spends up to ~30s on a richer prompt. Default value: `"balanced"`
  - Default: `"balanced"`
  - Examples: "disabled", "fast", "balanced", "quality"

- **`target_audio_url`** (`string`, _optional_):
  Optional URL of an audio clip at least 2 seconds long (maximum 15 MB) to pin to the generated soundtrack. Longer clips are trimmed to the generated video's length, keeping the beginning. The original audio replaces the output soundtrack, padded with silence if shorter than the video, without changing playback speed. Exceptionally high sample rates may be resampled to 96 kHz. Accepts an HTTP(S) URL or a base64 data URI.

- **`image_url`** (`string`, _optional_):
  Optional URL of the image to use as the first frame. When provided, the output canvas follows this image. If only end_image_url is provided, the canvas follows that last frame instead. If both images are omitted, the request is handled as text-to-video (16:9 by default).
  - Examples: "https://storage.googleapis.com/falserverless/example_inputs/hailuo23/pro_i2v_in.jpg"

- **`end_image_url`** (`string`, _optional_):
  Optional URL of the image to use as the last frame. It may be provided alone for end-only keyframe generation; in that case the output canvas follows this image.

- **`loras`** (`list<LoRAInput>`, _required_):
  List of LoRA adapters to apply to the transformer.
  - Array of LoRAInput



**Required Parameters Example**:

```json
{
  "prompt": "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain.",
  "loras": [
    {
      "path": "https://huggingface.co/owner/repo/resolve/main/lora.safetensors",
      "scale": 1
    }
  ]
}
```

**Full Example**:

```json
{
  "prompt": "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain.",
  "duration": 5,
  "resolution": "2K",
  "enable_safety_checker": true,
  "prompt_expansion_mode": "disabled",
  "image_url": "https://storage.googleapis.com/falserverless/example_inputs/hailuo23/pro_i2v_in.jpg",
  "loras": [
    {
      "path": "https://huggingface.co/owner/repo/resolve/main/lora.safetensors",
      "scale": 1
    }
  ]
}
```


### Output Schema

The API returns the following output format:

- **`video`** (`File`, _required_):
  The generated video

- **`expanded_prompt`** (`string`, _optional_):
  The prompt after expansion, as sent to the model. Null when prompt expansion was disabled, left the prompt unchanged, or was performed internally by MiniMax's hosted API.



**Example Response**:

```json
{
  "video": {
    "url": "",
    "content_type": "image/png",
    "file_name": "z9RV14K95DvU.png",
    "file_size": 4404019
  }
}
```


## Usage Examples

### cURL

```bash
curl --request POST \
  --url https://fal.run/minimax/h3/image-to-video/lora \
  --header "Authorization: Key $FAL_KEY" \
  --header "Content-Type: application/json" \
  --data '{
     "prompt": "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain.",
     "loras": [
       {
         "path": "https://huggingface.co/owner/repo/resolve/main/lora.safetensors",
         "scale": 1
       }
     ]
   }'
```

### Python

Ensure you have the Python client installed:

```bash
pip install fal-client
```

Then use the API client to make requests:

```python
import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
           print(log["message"])

result = fal_client.subscribe(
    "minimax/h3/image-to-video/lora",
    arguments={
        "prompt": "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain.",
        "loras": [{
            "path": "https://huggingface.co/owner/repo/resolve/main/lora.safetensors",
            "scale": 1
        }]
    },
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)
```

### JavaScript

Ensure you have the JavaScript client installed:

```bash
npm install --save @fal-ai/client
```

Then use the API client to make requests:

```javascript
import { fal } from "@fal-ai/client";

const result = await fal.subscribe("minimax/h3/image-to-video/lora", {
  input: {
    prompt: "The camera slowly pulls back from the scene, revealing the full landscape as clouds drift overhead and light shifts across the terrain.",
    loras: [{
      path: "https://huggingface.co/owner/repo/resolve/main/lora.safetensors",
      scale: 1
    }]
  },
  logs: true,
  onQueueUpdate: (update) => {
    if (update.status === "IN_PROGRESS") {
      update.logs.map((log) => log.message).forEach(console.log);
    }
  },
});
console.log(result.data);
console.log(result.requestId);
```


## Additional Resources

### Documentation

- [Model Playground](https://fal.ai/models/minimax/h3/image-to-video/lora)
- [API Documentation](https://fal.ai/models/minimax/h3/image-to-video/lora/api)
- [OpenAPI Schema](https://fal.ai/api/openapi/queue/openapi.json?endpoint_id=minimax/h3/image-to-video/lora)

### fal.ai Platform

- [Platform Documentation](https://fal.ai/docs/documentation)
- [Python Client](https://fal.ai/docs/api-reference/client-libraries/python)
- [JavaScript Client](https://fal.ai/docs/api-reference/client-libraries/javascript)

### Other agent-readable surfaces

This file covers one model. To find anything else:

- [Platform overview](https://fal.ai/llms.txt): Entry points and representative endpoint IDs
- [Documentation index](https://fal.ai/docs/llms.txt): Every documentation page
- [Full documentation text](https://fal.ai/docs/llms-full.txt): The whole documentation inlined
- Any other model: `https://fal.ai/models/<endpoint-id>/llms.txt`