Ollama MCP Server

This is a rebooted and actively maintained fork.
Original project: NightTrek/Ollama-mcp
This repository (hyzhak/ollama-mcp-server) is a fresh upstream with improved maintenance, metadata, and publishing automation.
See NightTrek/Ollama-mcp for project history and prior releases.

🚀 A powerful bridge between Ollama and the Model Context Protocol (MCP), enabling seamless integration of Ollama's local LLM capabilities into your MCP-powered applications.

🌟 Features

Complete Ollama Integration

Full API Coverage: Access all essential Ollama functionality through a clean MCP interface
OpenAI-Compatible Chat: Drop-in replacement for OpenAI's chat completion API
Local LLM Power: Run AI models locally with full control and privacy

Core Capabilities

🔄 Model Management
- Pull models from registries
- Push models to registries
- List available models
- Create custom models from Modelfiles
- Copy and remove models
🤖 Model Execution
- Run models with customizable prompts (response is returned only after completion; streaming is not supported in stdio mode)
- Vision/multimodal support: pass images to compatible models
- Chat completion API with system/user/assistant roles
- Configurable parameters (temperature, timeout)
- NEW: think parameter for advanced reasoning and transparency (see below)
- Raw mode support for direct responses
🛠 Server Control
- Start and manage Ollama server
- View detailed model information
- Error handling and timeout management

🚀 Quick Start

Prerequisites

Ollama installed on your system
Node.js (with npx, included with npm)

Configuration

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": "npx",
      "args": ["ollama-mcp-server"],
      "env": {
        "OLLAMA_HOST": "http://127.0.0.1:11434"  // Optional: customize Ollama API endpoint
      }
    }
  }
}

🛠 Developer Setup

Prerequisites

Ollama installed on your system
Node.js and npm

Installation

Install dependencies:

npm install

Build the server:

npm run build

🛠 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 quantum computing in simple terms"
  }
});

Run a Vision/Multimodal Model

// Run a model with an image (for vision/multimodal models)
await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "run",
  arguments: {
    name: "gemma3:4b",
    prompt: "Describe the contents of this image.",
    imagePath: "./path/to/image.jpg"
  }
});

Chat Completion (OpenAI-compatible)

await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "chat_completion",
  arguments: {
    model: "llama2",
    messages: [
      {
        role: "system",
        content: "You are a helpful assistant."
      },
      {
        role: "user",
        content: "What is the meaning of life?"
      }
    ],
    temperature: 0.7
  }
});

// Chat with images (for vision/multimodal models)
await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "chat_completion",
  arguments: {
    model: "gemma3:4b",
    messages: [
      {
        role: "system",
        content: "You are a helpful assistant."
      },
      {
        role: "user",
        content: "Describe the contents of this image.",
        images: ["./path/to/image.jpg"]
      }
    ]
  }
});

Note: The images field is optional and only supported by vision/multimodal models.

Create Custom Model

await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "create",
  arguments: {
    name: "custom-model",
    modelfile: "./path/to/Modelfile"
  }
});

🧠 Advanced Reasoning with the `think` Parameter

Both the run and chat_completion tools now support an optional think parameter:

think: true: Requests the model to provide step-by-step reasoning or "thought process" in addition to the final answer (if supported by the model).
think: false (default): Only the final answer is returned.

Example (`run` tool):

await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "run",
  arguments: {
    name: "deepseek-r1:32b",
    prompt: "how many r's are in strawberry?",
    think: true
  }
});

If the model supports it, the response will include a :::thinking ... ::: block with detailed reasoning before the final answer.

Example (`chat_completion` tool):

await mcp.use_mcp_tool({
  server_name: "ollama",
  tool_name: "chat_completion",
  arguments: {
    model: "deepseek-r1:32b",
    messages: [
      { role: "user", content: "how many r's are in strawberry?" }
    ],
    think: true
  }
});

The model's reasoning (if provided) will be included in the message content.

Note: Not all models support the think parameter. Advanced models (e.g., "deepseek-r1:32b", "magistral") may provide more detailed and accurate reasoning when think is enabled.

🔧 Advanced Configuration

OLLAMA_HOST: Configure custom Ollama API endpoint (default: http://127.0.0.1:11434)
Timeout settings for model execution (default: 60 seconds)
Temperature control for response randomness (0-2 range)

🤝 Contributing

Contributions are welcome! Feel free to:

Report bugs
Suggest new features
Submit pull requests

📝 License

MIT License - feel free to use in your own projects!

Built with ❤️ for the MCP ecosystem

Deploy Server

HTTP connection URL

security – no known vulnerabilities

license - permissive license

quality - confirmed to work

How are these scores calculated?

local-only server

The server can only run on the client's local machine because it depends on local resources.

Tools

View all tools

A bridge that integrates Ollama's local LLM capabilities into MCP-powered applications, enabling users to run, manage, and interact with AI models locally with full control and privacy.

Related MCP Servers

MCP Ollama Server
emgeee
A
security
A
license
A
quality
MCP Ollama server integrates Ollama models with MCP clients, allowing users to list models, get detailed information, and interact with them through questions.
Last updated -
3
27
MIT License
Ollama MCP Server
rawveg
-
security
A
license
-
quality
Enables seamless integration between Ollama's local LLM models and MCP-compatible applications, supporting model management and chat interactions.
Last updated -
69
98
AGPL 3.0
Ollama MCP Server
NightTrek
A
security
F
license
A
quality
A bridge that enables seamless integration of Ollama's local LLM capabilities into MCP-powered applications, allowing users to manage and run AI models locally with full API coverage.
Last updated -
10
73
MCP Base
jsmiff
-
security
F
license
-
quality
A generic Model Context Protocol framework for building AI-powered applications that provides standardized ways to create MCP servers and clients for integrating LLMs with support for Ollama and Supabase.
Last updated -

View all related MCP servers

Ollama MCP Server