# MiniMax H3 Reference to Video

This node generates a video using the MiniMax H3 models, conditioned on reference images, videos, and audio. References are referred to in the prompt by their connection order: "@Image1", "@Image2", "@Video1", "@Audio1", and so on. Two models are available: "MiniMax H3" and "MiniMax H3 Max".

## Inputs

### Common Inputs

| Parameter | Description | Data Type | Required | Range |
|-----------|-------------|-----------|----------|-------|
| `model` | Model to use for video generation (default: "MiniMax H3"). Selecting "MiniMax H3" provides the MiniMax H3 generation and reference inputs below. Selecting "MiniMax H3 Max" provides the MiniMax H3 Max generation and reference inputs below. | DYNAMIC_COMBO | Yes | "MiniMax H3"<br>"MiniMax H3 Max" |
| `seed` | Random seed. The same request with the same seed gives similar, but not guaranteed identical, results (default: 42). | INT | Yes | 0 to 4294967295 |
| `watermark` | Whether to add an AIGC watermark to the video (default: false). Only supported by the MiniMax H3 model. | BOOLEAN | No | true<br>false |

### MiniMax H3 Inputs

These inputs are available when "MiniMax H3" is selected as the model.

| Parameter | Description | Data Type | Required | Range |
|-----------|-------------|-----------|----------|-------|
| `prompt` | Text prompt for video generation. Reference media can be referred to by their order, for example "@Image1", "@Image2", "@Video1", or "@Audio1". | STRING | Yes | Minimum 1 character |
| `resolution` | Resolution of the output video (default: "768P"). | COMBO | Yes | "768P"<br>"2K" |
| `ratio` | Aspect ratio of the output video (default: "adaptive"). | COMBO | Yes | "adaptive"<br>"16:9"<br>"4:3"<br>"1:1"<br>"3:4"<br>"9:16"<br>"21:9" |
| `duration` | Duration of the output video in seconds (default: 5). | INT | Yes | 4 to 15 |

### MiniMax H3 Max Inputs

These inputs are available when "MiniMax H3 Max" is selected as the model.

| Parameter | Description | Data Type | Required | Range |
|-----------|-------------|-----------|----------|-------|
| `prompt` | Text prompt for video generation. Reference media can be referred to by their order, for example "@Image1", "@Image2", "@Video1", or "@Audio1". | STRING | Yes | 1 to 50000 characters |
| `resolution` | Resolution of the output video (default: "768P"). | COMBO | Yes | "480P"<br>"768P" |
| `ratio` | Aspect ratio of the output video (default: "adaptive"). | COMBO | Yes | "adaptive"<br>"16:9"<br>"4:3"<br>"1:1"<br>"3:4"<br>"9:16"<br>"21:9" |
| `duration` | Duration of the output video in seconds (default: 5). | INT | Yes | 5 to 15 |
| `prompt_expansion_mode` | How much effort is spent rewriting the prompt before generation (default: "balanced"). | COMBO | Yes | "balanced"<br>"quality" |
| `reference_detail` | Detail level at which reference images are sent. "high" sends them at the largest size the model uses (up to a 2048 pixel short side); "standard" downsizes them to at most 2048x1024 to reduce the reference cost (default: "standard"). | COMBO | Yes | "high"<br>"standard" |

### Reference Inputs

These reference inputs are shared by both models. Each is a growable slot.

| Parameter | Description | Data Type | Required | Range |
|-----------|-------------|-----------|----------|-------|
| `reference_images` | Growable slot: connect up to 9 items (`image_1`...`image_9`). Subject or style reference images, referred to in the prompt as "@Image1".."@Image9" in connection order. Up to 9 images. | IMAGE | No | 0 to 9 images |
| `reference_videos` | Growable slot: connect up to 3 items (`video_1`...`video_3`). Motion or scene reference videos, referred to in the prompt as "@Video1".."@Video3" in connection order. Up to 3 videos, 2-15 seconds each, 15 seconds in total. | VIDEO | No | 0 to 3 videos |
| `reference_audios` | Growable slot: connect up to 3 items (`audio_1`...`audio_3`). Audio references, referred to in the prompt as "@Audio1".."@Audio3" in connection order. Up to 3 clips, 2-15 seconds each, 15 seconds in total. Cannot be used without a reference image or video. | AUDIO | No | 0 to 3 clips |

### Parameter Constraints

- At least one reference image or one reference video is required. Reference audio alone is not accepted.
- Each reference image must have an aspect ratio between approximately 0.4 and 2.5 (2:5 to 5:2) and a minimum width and height of 256 pixels.
- Each reference video must be between 2 and 15 seconds long with a frame rate between 23.976 and 60 FPS. The total duration of all reference videos cannot exceed 15 seconds.
- Each reference audio clip must be between 2 and 15 seconds long. The total duration of all reference audio clips cannot exceed 15 seconds.
- When "MiniMax H3 Max" is selected, the `watermark` setting must be disabled.
- When "MiniMax H3 Max" is selected, the total number of reference files (images, videos, and audio combined) cannot exceed 12.

## Outputs

| Output Name | Description | Data Type |
|-------------|-------------|-----------|
| `video` | The generated video. | VIDEO |

> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/MinimaxHailuo03ReferenceNode/en.md)

---
**Source fingerprint (SHA-256):** `4c0dfb349809b51393b8ec8afd235977f1822c0c8ae6dc42a2c264271bb50b53`
