Siliconflow Flux MCP Server

🚀 SiliconFlow Flux MCP Server

This is a server based on the Model Context Protocol (MCP) that allows AI to use SiliconFlow's Flux model for image generation via the MCP protocol.

🚀 Quick Start

The SiliconFlow Flux MCP Server is a powerful tool that enables AI to generate high - quality images using SiliconFlow's Flux model through the MCP protocol.

✨ Features

• Allows AI assistants to call SiliconFlow's Flux model via the MCP protocol to generate high - quality images.
• Supports multiple image resolutions: 1024x1024, 960x1280, 768x1024, 720x1440, 720x1280.
• Allows customization of generation parameters, such as random seeds.
• Caches recent image generation results.

📦 Installation

System Requirements

• Node.js v18.0.0 or higher.

You can verify the Node.js installation with the following command:

node --version  # Should display v18.0.0 or higher

Installation Steps

1. Clone the repository:

git clone https://github.com/lioensky/siliconflow-flux-mcp-server.git

2. Install dependencies:

npm install

3. Build the project:

npm run build

📚 Documentation

API Key Configuration

Before use, you need to:

1. Create or edit the .env file in the project root directory.
2. Obtain a 14 - yuan quota key via https://cloud.siliconflow.cn/i/HxM1olJu.
3. Add your SiliconFlow API key:

SILICONFLOW_API_KEY=Your_API_Key

⚠️ Important Note

You need to provide your own SiliconFlow API key to use this service.

MCP Server Configuration

Add the following content to your MCP configuration file:

{
  "mcpServers": {
    "siliconflow-flux-mcp": {
      "command": "node",
      "args": ["Path/to/siliconflow-flux-mcp-server/build/index.js"],
      "env": {
        "SILICONFLOW_API_KEY": "Your_API_Key"
      }
    }
  }
}

💻 Usage Examples

Basic Usage

After configuration, AI assistants can call this service to generate images via the MCP protocol. The service provides a generate_image tool that accepts the following parameters:

• prompt: Image generation prompt (English is recommended for the best results).
• resolution: Image resolution, supporting multiple standard sizes.
• seed (optional): Random seed for generating reproducible results.

Advanced Usage

You can use the MCP Inspector tool to test the server:

npm run inspector

📄 License

Please refer to the license file in the project repository.