Doubao Image Video MCP

🚀 Doubao Image/Video Generation MCP Server

An MCP (Model Context Protocol) server supporting Doubao image and video generation.

🚀 Quick Start

This is an MCP server that supports Doubao image and video generation. It enables users to generate high - quality images and videos through the provided API.

✨ Features

• Image Generation: Generate high - quality images using the Doubao Seedream model.
• Video Generation: Generate videos using the Doubao Seedance model.
• Task Query: Query the status of video generation tasks.

📦 Installation

Local Installation

cd doubao-image-video-mcp
pnpm install
pnpm build

💻 Usage Examples

Get API Key

1. Visit the Volcengine Doubao Large Model Platform.
2. Register/Log in to your account.
3. Obtain the API Key in the console.

Create an Inference Access Point

1. Visit the Volcengine Console.
2. Go to the "Activation Management" page and participate in the "Collaboration Reward Program" to enjoy a free resource package.
3. Authorize the image generation access point (Doubao - Seedream - 4.5 model).
4. Authorize the video generation access point (Doubao - Seedance - 1.0 - lite - t2v model).
5. Go to the "Security Management" page, click on content security, and copy the access point ID to the configuration file.

PS: Different access points are configured for image generation and video generation.

Configuration

Use in Claude Code (Recommended)

Add the following to the Claude Code configuration file:

• macOS/Linux: ~/.claude/config.json
• Windows: %APPDATA%\claude\config.json

{
  "mcpServers": {
    "doubao-giv": {
      "command": "npx",
      "args": ["-y", "doubao-image-video-mcp@latest"],
      "env": {
        "DOUBAO_API_KEY": "your_api_key_here",
        "DOUBAO_IMAGE_ENDPOINT_ID": "ep-20241227-xxxxxxxxxxxxx",
        "DOUBAO_VIDEO_ENDPOINT_ID": "ep-20241227-xxxxxxxxxxxxx"
      }
    }
  }
}

General MCP Client Configuration

{
  "mcpServers": {
    "doubao-giv": {
      "command": "npx",
      "args": ["-y", "doubao-image-video-mcp@latest"],
      "env": {
        "DOUBAO_API_KEY": "your_api_key_here",
        "DOUBAO_IMAGE_ENDPOINT_ID": "ep-20241227-xxxxxxxxxxxxx",
        "DOUBAO_VIDEO_ENDPOINT_ID": "ep-20241227-xxxxxxxxxxxxx"
      }
    }
  }
}

Using npx -y can automatically install the latest version without manually managing local files.

Local Development Configuration

If running from source code:

{
  "mcpServers": {
    "doubao-giv": {
      "command": "node",
      "args": ["/absolute/path/to/doubao-image-video-mcp/dist/index.js"],
      "env": {
        "DOUBAO_API_KEY": "your_api_key_here",
        "DOUBAO_IMAGE_ENDPOINT_ID": "ep-20241227-xxxxxxxxxxxxx",
        "DOUBAO_VIDEO_ENDPOINT_ID": "ep-20241227-xxxxxxxxxxxxx"
      }
    }
  }
}

Environment Variables

⚠️ Important Note

You must create an inference access point in the Volcengine console and configure the access point ID. Directly using the model name may cause call failures due to permission issues.

Supported Tools

1. generate_image

Generate a single image.

Parameters:

• prompt (Required): Text description of the image.
• model (Optional): Model selection, default is doubao-seedream-4-5.
• size (Optional): Image size, default is 2560x1440.
• image_url (Optional): URL of the reference image (image - to - image).
• ref_image_urls (Optional): Array of URLs of multiple reference images (multi - image fusion).
• req_key (Optional): Request identifier.

2. generate_video

Generate a video (asynchronous task).

Parameters:

• prompt (Required): Text description of the video (maximum 500 characters).
• model (Optional): Model selection, default is doubao-seedance-1.0-lite-t2v.
• video_duration (Optional): Video duration (seconds), default is 5.
• fps (Optional): Frame rate, default is 24.
• resolution (Optional): Resolution, default is 1080p.
• first_frame_image_url (Optional): URL of the first - frame image (image - to - video).
• ref_image_urls (Optional): Array of URLs of reference images.
• req_key (Optional): Request identifier.

Return: Task ID. You need to use query_video_task to query the result.

3. query_video_task

Query the status of a video generation task.

Parameters:

• task_id (Required): Video generation task ID.

Return: Task status and result.

Supported Models

Image Generation Models

Video Generation Models

📚 Documentation

• Create Video Generation Task API
• Query Video Generation Task API
• Image Generation API

📄 License

MIT

Frequently Asked Questions

Q: How long does video generation take?

A: Video generation is an asynchronous task, usually taking from a few seconds to a few minutes, depending on the video duration and model selection.

Q: How can I get the inference access point Endpoint ID?

A: Obtain it after creating an inference access point in the Volcengine console. Using the endpoint_id parameter is more stable than directly using the model name.

Q: What image sizes are supported?

A: Image generation supports various sizes. It is recommended to use high - resolution sizes such as 1920x2160, 1920x2560, 2160x3840 (the Doubao API requires images to have at least 3686400 pixels).

Contribute

We welcome you to submit Issues and Pull Requests!

1. Fork this repository.
2. Create a feature branch (git checkout -b feature/AmazingFeature).
3. Commit your changes (git commit -m 'Add some AmazingFeature').
4. Push to the branch (git push origin feature/AmazingFeature).
5. Open a Pull Request.

Contribution Guidelines

• Follow the existing code style.
• Add necessary tests and documentation.
• Run pnpm build before submission to ensure successful building.

Release Process

The project uses GitHub Actions to automatically publish to npm:

1. Update the version number in package.json.
2. Create and push a Git tag:

   git tag v1.0.0
   git push origin v1.0.0
   

3. GitHub Actions will automatically build and publish to npm.

Star History

💡 Usage Tip

If this project is helpful to you, please give it a Star ⭐️.