home / mcp / deepseek mcp server

Deepseek MCP Server

MCP Server for deepseek integration in Claude desktop

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "vincentf305-mcp-server-deepseek": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DEEPSEEK_API_KEY",
        "mcp_server_deepseek"
      ],
      "env": {
        "DEEPSEEK_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

This MCP server lets Claude Desktop connect to and use Deepseek models running inside Docker or natively via Python. It exposes a managed MCP endpoint that Claude Desktop can query to run Deepseek models, enabling seamless model execution within your Claude workflow.

How to use

You connect to the Deepseek MCP server from Claude Desktop by configuring an MCP server entry. You can run the server either as a Docker container or locally with Python. Once the server is running, Claude Desktop will communicate with it to send model control requests and receive results.

How to install

Prerequisites you need before starting:

- Docker installed on your machine

- Python 3.11 or later

- A Deepseek API key

- Claude Desktop installed and ready for MCP configuration

Using Docker to run the MCP server

# Build the Docker image
docker build -t mcp_server_deepseek .

# Run the container (exposes MCP on port 8765 and passes your API key)
docker run -d \
  --name mcp-server-deepseek \
  -p 8765:8765 \
  -e DEEPSEEK_API_KEY=your_api_key_here \
  mcp-server-deepseek

Running the MCP server locally (Python)

python -m mcp_server_deepseek.server

Configure Claude Desktop to connect to the Deepseek MCP server

{
  "mcpServers": {
    "deepseek": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DEEPSEEK_API_KEY",
        "mcp_server_deepseek"
      ],
      "env": {
        "DEEPSEEK_API_KEY": "your_api_key_here"
      }
    }
  }
}

Notes on setup and environment variables

- You can store your API key in a local environment variable file and reference it in the container run command, or replace the placeholder with your actual Deepseek API key directly in the command.

Troubleshooting and tips

- If Claude Desktop cannot connect, verify that port 8765 on the host is accessible and that the Docker container is running. Check the container logs for errors related to authentication or missing dependencies.

Security considerations

- Keep your Deepseek API key secure. Do not commit it to version control. Use environment variables or secret management where possible.