Deepseek R1 MCP Server

A Model Context Protocol (MCP) server implementation for the Deepseek R1 language model. Deepseek R1 is a powerful language model optimized for reasoning tasks with a context window of 8192 tokens.

Why Node.js? This implementation uses Node.js/TypeScript as it provides the most stable integration with MCP servers. The Node.js SDK offers better type safety, error handling, and compatibility with Claude Desktop.

Quick Start

Installing via Smithery

To install Deepseek R1 for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @66julienmartin/mcp-server-deepseek_r1 --client claude

Installing manually

# Clone and install
git clone https://github.com/66julienmartin/MCP-server-Deepseek_R1.git
cd deepseek-r1-mcp
npm install

# Set up environment
cp .env.example .env  # Then add your API key

# Build and run
npm run build

Prerequisites

• Node.js (v18 or higher)
• npm
• Claude Desktop
• Deepseek API key

Model Selection

By default, this server uses the deepseek-R1 model. If you want to use DeepSeek-V3 instead, modify the model name in src/index.ts:

// For DeepSeek-R1 (default)
model: "deepseek-reasoner"

// For DeepSeek-V3
model: "deepseek-chat"

Project Structure

deepseek-r1-mcp/
├── src/
│   ├── index.ts             # Main server implementation
├── build/                   # Compiled files
│   ├── index.js
├── LICENSE
├── README.md
├── package.json
├── package-lock.json
└── tsconfig.json

Configuration

1. Create a .env file:

DEEPSEEK_API_KEY=your-api-key-here

2. Update Claude Desktop configuration:

{
  "mcpServers": {
    "deepseek_r1": {
      "command": "node",
      "args": ["/path/to/deepseek-r1-mcp/build/index.js"],
      "env": {
        "DEEPSEEK_API_KEY": "your-api-key"
      }
    }
  }
}

Development

npm run dev     # Watch mode
npm run build   # Build for production

Features

• Advanced text generation with Deepseek R1 (8192 token context window)
• Configurable parameters (max_tokens, temperature)
• Robust error handling with detailed error messages
• Full MCP protocol support
• Claude Desktop integration
• Support for both DeepSeek-R1 and DeepSeek-V3 models

API Usage

{
  "name": "deepseek_r1",
  "arguments": {
    "prompt": "Your prompt here",
    "max_tokens": 8192,    // Maximum tokens to generate
    "temperature": 0.2     // Controls randomness
  }
}

The Temperature Parameter

The default value of temperature is 0.2.

Deepseek recommends setting the temperature according to your specific use case:

Error Handling

The server provides detailed error messages for common issues:

• API authentication errors
• Invalid parameters
• Rate limiting
• Network issues

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT