# Seedance-2.5 Multimodal Reference To Video 

> AI model for generating video content

## Overview

- **Model ID**: `seedance-2-5-multi-reference-video`
- **Category**: video
- **Provider**: byteplus
- **Source Type**: Closed Source Model
- **Status**: model_ready
- **Screenshot**: `https://assets.modelslab.ai/generations/7aedebf2-9038-4312-80d9-c6d6040a539e.webp`

## API Information

This model can be used via our HTTP API. See the API documentation and usage examples below.

### Endpoint

- **URL**: `https://modelslab.com/api/v7/video-fusion/image-to-video`
- **Method**: POST

### Parameters

- **`init_image`** (required): Upload upto 30 reference image
  - Type: array

- **`init_audio`** (required): Upload upto 10 Reference Audio 
  - Type: array

- **`init_video`** (required): 
  - Type: array

- **`prompt`** (required): Text description of the video you want to generate
  - Type: textarea
  - Example: Enter prompt for video

- **`aspect_ratio`** (required): 
  - Type: select (options: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16)

- **`resolution`** (required): 
  - Type: select (options: 720P, 480P)

- **`duration`** (required): Choose the duration between 4-15 seconds
  - Type: number (range: 5-30)

## Usage Examples

### cURL

```bash
curl --request POST \
  --url https://modelslab.com/api/v7/video-fusion/image-to-video \
  --header "Content-Type: application/json" \
  --data '{
    "key": "YOUR_API_KEY",
    "model_id": "seedance-2-5-multi-reference-video",
    "prompt": "A cinematic brand concept short film. 

@Image1
 is the first frame. The image shakes slightly as the camera gradually pushes in toward the tree shadows rapidly receding outside the window. The tree shadows retreat faster and faster, then suddenly cut to 

@Image2
. The speed abruptly slows, and the camera moves gently forward along the stream amid birdsong and flowers. The camera tilts downward and enters the underwater world. The sound effect includes underwater bubbles. A group of orange jellyfish gracefully swims past the camera, 

@Image3
. The camera slowly pulls back. A school of small fish drifts past the lens, then passes from the water into the interior through the window, 

@Image4
. The girl looks left and right, watching the fish. The camera slowly pulls back, the image goes out of focus, then refocuses and becomes clear again. Following the rhythm of the music, it cuts through: a Chinese garden lattice window",
    "aspect_ratio": "21:9",
    "resolution": "720p",
    "duration": "5"
  }'
```

### Python

```python
import requests

response = requests.post(
    "https://modelslab.com/api/v7/video-fusion/image-to-video",
    headers={
        "Content-Type": "application/json"
    },
    json={
        "key": "YOUR_API_KEY",
        "model_id": "seedance-2-5-multi-reference-video",
        "prompt": "A cinematic brand concept short film. 

@Image1
 is the first frame. The image shakes slightly as the camera gradually pushes in toward the tree shadows rapidly receding outside the window. The tree shadows retreat faster and faster, then suddenly cut to 

@Image2
. The speed abruptly slows, and the camera moves gently forward along the stream amid birdsong and flowers. The camera tilts downward and enters the underwater world. The sound effect includes underwater bubbles. A group of orange jellyfish gracefully swims past the camera, 

@Image3
. The camera slowly pulls back. A school of small fish drifts past the lens, then passes from the water into the interior through the window, 

@Image4
. The girl looks left and right, watching the fish. The camera slowly pulls back, the image goes out of focus, then refocuses and becomes clear again. Following the rhythm of the music, it cuts through: a Chinese garden lattice window",
        "aspect_ratio": "21:9",
        "resolution": "720p",
        "duration": "5"
    }
)

print(response.json())
```

### JavaScript

```javascript
fetch("https://modelslab.com/api/v7/video-fusion/image-to-video", {
  method: "POST",
  headers: {
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    "key": "YOUR_API_KEY",
    "model_id": "seedance-2-5-multi-reference-video",
    "prompt": "A cinematic brand concept short film. 

@Image1
 is the first frame. The image shakes slightly as the camera gradually pushes in toward the tree shadows rapidly receding outside the window. The tree shadows retreat faster and faster, then suddenly cut to 

@Image2
. The speed abruptly slows, and the camera moves gently forward along the stream amid birdsong and flowers. The camera tilts downward and enters the underwater world. The sound effect includes underwater bubbles. A group of orange jellyfish gracefully swims past the camera, 

@Image3
. The camera slowly pulls back. A school of small fish drifts past the lens, then passes from the water into the interior through the window, 

@Image4
. The girl looks left and right, watching the fish. The camera slowly pulls back, the image goes out of focus, then refocuses and becomes clear again. Following the rhythm of the music, it cuts through: a Chinese garden lattice window",
    "aspect_ratio": "21:9",
    "resolution": "720p",
    "duration": "5"
  })
})
.then(response => response.json())
.then(data => console.log(data));
```

## Integration Options

### CLI

Install: `curl -fsSL https://modelslab.sh/install.sh | sh` or `brew install modelslab/tap/modelslab`

```bash
modelslab auth login                                    # Authenticate
modelslab models search "seedance-2-5-multi-reference-video"                       # Find this model
modelslab models detail --id seedance-2-5-multi-reference-video                    # Get model details
modelslab generate image --prompt "..." --model seedance-2-5-multi-reference-video  # Generate with model
modelslab config set generation.default_model seedance-2-5-multi-reference-video   # Set as default
```

- Website: https://modelslab.sh
- GitHub: https://github.com/ModelsLab/modelslab-cli

### MCP Servers (Model Context Protocol)

For Claude Code, Cursor, VS Code, Windsurf, and any MCP-compatible agent.

- Generation: `https://modelslab.com/mcp/v7` (API key auth, 23 tools)
- Agent Control Plane: `https://modelslab.com/mcp/agents` (Bearer token auth, 10 tools)

Claude Code config (`~/.claude/settings.json`):

```json
{
  "mcpServers": {
    "modelslab-v7": {
      "url": "https://modelslab.com/mcp/v7",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

- Documentation: https://docs.modelslab.com/mcp-web-api/overview

### Agent Skills

Install skill files directly into AI coding agents:

```bash
npx skills add modelslab/skills --all                          # All skills
npx skills add modelslab/skills --skill image-generation       # Specific skill
npx skills add modelslab/skills --all -a claude-code -a cursor # Target agents
```

- GitHub: https://github.com/ModelsLab/skills
- Documentation: https://docs.modelslab.com/agent-skills

### SDKs

- Python: `pip install modelslab`
- TypeScript: `npm install modelslab`
- PHP: `composer require modelslab/modelslab`
- Go: `go get github.com/modelslab/modelslab-go`
- Dart: `dart pub add modelslab`

## Links

- [Model Playground](https://modelslab.com/models/byteplus/seedance-2-5-multi-reference-video)
- [API Documentation](https://docs.modelslab.com)
- [ModelsLab Platform](https://modelslab.com)