muhammadsohaib56/MCP-Server-Implementation-with-OLLAMA-LLM-

MCP Unit Converter with LLM Integration

The **MCP Unit Converter** is a Python-based application that integrates the [MCP SDK](https://github.com/modelcontextprotocol) with the **Llama 3.1 8B** model (via [Ollama](https://ollama.com)) to perform **precise unit conversions** for length, mass, and temperature.  

The system follows a **client-server architecture**:
- **client.py** → Interacts with the user and the LLM.
- **server.py** → Handles conversion logic using the MCP `convert` tool.
- **units://supported_units.json** → Defines all supported units.

---

## ✨ Features

- **Unit Conversions**  
  - Length: `mm, cm, m, km, in, ft, yd, mi`  
  - Mass: `mg, g, kg, tonne, oz, lb`  
  - Temperature: `C, F, K`

- **LLM Integration**  
  Natural language input like:  

convert 100 cm to inches convert 32 fahrenheit to celsius

- **MCP SDK**  
Uses MCP tools for precise calculations instead of relying only on the LLM.

- **Interactive CLI**  
Command-line interface with `exit` command to quit.

- **Precision Control**  
Default: 6 decimal places (configurable).

---

## 📂 Project Structure

---

## 🔧 Prerequisites

- **Python** ≥ 3.10  
- **Ollama** installed and running with **Llama 3.1 8B** model  
- **Operating System**: Windows/Linux/macOS  
- **Disk Space**: ~15GB free (Ollama model files + temp files)  
- **RAM**: 8GB (16GB recommended)  

---

## ⚙️ Installation

1. **Clone the Repository**  
   ```bash
   git clone https://github.com/your-username/mcp-unit-converter-llm.git
   cd mcp-unit-converter-llm

2. Install Dependencies

   pip install -r requirements.txt
   

3. Install Ollama

   • Download Ollama from ollama.com

   • Pull the Llama 3.1 8B model:

     ollama pull llama3.1:8b
     

🚀 Usage

1. Start Ollama Server

   ollama serve
   

   Verify:

   curl http://127.0.0.1:11434
   

   Should return: Ollama is running.

2. Run the Client

   python client.py
   

3. Example Queries

   You: convert 100 cm to inches
   Assistant: The conversion of 100 cm to inches is 39.37 inches.
   

   You: convert 32 fahrenheit to celsius
   Assistant: The conversion of 32 Fahrenheit to Celsius is 0.0 °C.
   

   Type exit to quit.

⚙️ How It Works

• Client (client.py)

  • Starts the MCP server (server.py)
  • Retrieves supported units
  • Uses ollama.chat for LLM interaction
  • Handles tool calls (convert) and displays results

• Server (server.py)

  • Defines the convert tool
  • Uses conversion factors & formulas from supported_units.json
  • Serves resources via MCP

• LLM Integration

  • Llama 3.1 8B interprets queries
  • Calls the MCP tool when conversions are required

🧾 Supported Units

• Length: mm, cm, m, km, in, ft, yd, mi
• Mass: mg, g, kg, tonne, oz, lb
• Temperature: Celsius (C), Fahrenheit (F), Kelvin (K)

🛠 Troubleshooting

• Ollama not running → Run ollama serve and check port 11434.

• Low disk space → Ensure at least 15GB free.

• No tool calls → Check server.py and ensure convert tool is correctly registered.

• Slow performance → Try smaller model:

  ollama pull llama3:8b
  

🤝 Contributing

Contributions are welcome!

1. Fork the repo
2. Create a feature branch (git checkout -b feature/YourFeature)
3. Commit changes (git commit -m "Add YourFeature")
4. Push (git push origin feature/YourFeature)
5. Open a Pull Request

📜 License

This project is licensed under the MIT License.

Would you like me to also generate a sample requirements.txt (with exact package names like mcp and ollama) so your repo is fully ready-to-push?