Ollama MCP

🚀 Ollama MCP Server

A powerful bridge connecting Ollama and the Model Context Protocol (MCP), enabling you to seamlessly integrate Ollama's local large language model capabilities into your MCP applications.

🚀 Quick Start

Prerequisites

• Ollama installed on your system.
• Node.js and npm/pnpm environment configured.

Installation Steps

1. Install dependencies:

pnpm install

2. Build the server:

pnpm run build

Configuration Guide

Add the server to your MCP configuration:

For Claude Desktop:

MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "ollama": {
      "command": "node",
      "args": ["/path/to/ollama-server/build/index.js"],
      "env": {
        "OLLAMA_HOST": "http://127.0.0.1:11434"  // Optional: Customize the Ollama API endpoint
      }
    }
  }
}

✨ Features

Full Ollama Integration

• Comprehensive API Support: Access all key Ollama features through a simple MCP interface.
• OpenAI-compatible Chat Function: Directly replace OpenAI's chat completion API.
• Local Model Control: Run AI models locally with full control and privacy protection.

Core Capabilities

• 🔄 Model Management
  • Pull models from the registry.
  • Push models to the registry.
  • List available models.
  • Create custom models using Modelfile.
  • Copy and delete models.
• 🤖 Model Execution
  • Run models with customizable prompts.
  • Chat completion API with system/user/assistant roles.
  • Configurable parameters (temperature, timeout).
  • Support for raw mode with direct responses.
• 🛠 Server Control
  • Start and manage the Ollama server.
  • View detailed model information.
  • Error handling and timeout management.

💻 Usage Examples

Pull and Run a Model

// Pull a model
await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "pull",
  arguments: {
    name: "llama2"
  }
});

// Run the model
await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "run",
  arguments: {
    name: "llama2",
    prompt: "Explain the simple principles of quantum computing"
  }
});

Chat Completion (OpenAI-compatible)

await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "chat_completion",
  arguments: {
    messages: [
      { role: "user", content: "Explain the simple principles of quantum computing" }
    ],
    temperature: 0.7
  }
});

Create a Custom Model

await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "create_model",
  arguments: {
    name: "my_custom_model",
    source: "local://path/to/my/model.bin"
  }
});

📚 Documentation

• Ollama Official Documentation
• MCP Protocol Specification

📄 Tips

⚠️ Important Note

Ensure that your Ollama server is correctly configured and running.

💡 Usage Tip

When using environment variables, make sure they are correctly set to avoid connection issues. For advanced use cases, refer to the detailed documentation of the MCP protocol.

Let's explore the powerful capabilities of local AI together!