deepseek-mcp-server  by DMontgomery40

MCP server for DeepSeek API integration

created 6 months ago
261 stars

Top 98.0% on sourcepulse

GitHubView on GitHub
Project Summary

This project provides a Model Context Protocol (MCP) server for DeepSeek's advanced language models, enabling seamless integration with MCP-compatible applications like Claude Desktop. It acts as a proxy, allowing users to leverage DeepSeek's powerful AI capabilities while potentially anonymizing their API interactions.

How It Works

The server functions as a bridge, translating requests from MCP-compatible clients into calls to the DeepSeek API. It supports dynamic configuration of model parameters such as temperature, max tokens, and sampling methods. A key feature is its automatic model fallback mechanism, switching to a more general-purpose model if the primary reasoning model is unavailable, and allowing manual model switching via natural language prompts.

Quick Start & Requirements

  • Installation: npm install -g deepseek-mcp-server or via Smithery: npx -y @smithery/cli install @dmontgomery40/deepseek-mcp-server --client claude
  • Prerequisites: Node.js, npm, DeepSeek API Key.
  • Configuration: Requires adding server details to claude_desktop_config.json, including the DEEPSEEK_API_KEY.
  • Testing: Can be tested locally using npx @modelcontextprotocol/inspector node ./build/index.js after building the server with npm run build.

Highlighted Details

  • Supports multi-turn conversation history and context preservation.
  • Enables resource discovery for available models and configurations.
  • Offers fine-grained control over model parameters like temperature, max tokens, top P, presence, and frequency penalties.
  • Features automatic fallback to deepseek-chat (v3) if deepseek-reasoner (R1) is unavailable.

Maintenance & Community

The project is maintained by DMontgomery40. No specific community channels or roadmap details are provided in the README.

Licensing & Compatibility

The project is released under the MIT license, permitting commercial use and integration with closed-source applications.

Limitations & Caveats

The server defaults to the R1 model (deepseek-reasoner) but can be manually switched to deepseek-chat (v3). The README does not specify any known bugs or unsupported platforms.

Health Check
Last commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
1
Issues (30d)
0
Star History
43 stars in the last 90 days

Explore Similar Projects

Starred by Georgios Konstantopoulos Georgios Konstantopoulos(CTO, General Partner at Paradigm) and Jiayi Pan Jiayi Pan(Author of SWE-Gym; AI Researcher at UC Berkeley).

DeepSeek-V2 by deepseek-ai

0.1%
5k
MoE language model for research/API use
created 1 year ago
updated 10 months ago
Starred by Michael Han Michael Han(Cofounder of Unsloth), Sebastian Raschka Sebastian Raschka(Author of Build a Large Language Model From Scratch), and
6 more.

DeepSeek-R1 by deepseek-ai

0.1%
91k
Reasoning models research paper
created 6 months ago
updated 1 month ago
Feedback? Help us improve.