Claude Lmstudio Bridge

🚀 Claude-LMStudio Bridge Tool

A MCP server that connects Claude to the LLM running locally in LM Studio, enabling Claude to access your local LLM.

📋 Overview

This tool allows Claude to access your locally running LLM in LM Studio and provides the following features:

• List all available models in LM Studio.
• Generate text using your local LLM.
• Support conversation completion through local models.
• A health check tool to verify the connection status with LM Studio.

📋 Prerequisites

• Install and run Claude Desktop with MCP support.
• Install and run LM Studio and enable the API server.
• Install Python version 3.8 or higher.

🚀 Quick Start (Recommended)

For macOS/Linux:

1. Clone the repository

git clone https://github.com/infinitimeless/claude-lmstudio-bridge.git
cd claude-lmstudio-bridge

2. Run the installation script

chmod +x setup.sh
./setup.sh

3. Configure Claude Desktop as instructed by the installation script.

For Windows:

1. Clone the repository

git clone https://github.com/infinitimeless/claude-lmstudio-bridge.git
cd claude-lmstudio-bridge

2. Run the installation script

setup.bat

3. Configure Claude Desktop as instructed by the installation script.

🛠️ Manual Installation

If you prefer to set up manually:

1. Create a virtual environment (optional but recommended)

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

2. Install the required packages

pip install -r requirements.txt

3. Configure Claude Desktop:
   • Open Claude Desktop preferences.
   • Navigate to the "MCP Server" section.
   • Add a new MCP server with the following configuration:
     • Name: lmstudio-bridge
     • Command: /bin/bash (macOS/Linux) or cmd.exe (Windows)
     • Arguments:
       • macOS/Linux: /path/to/claude-lmstudio-bridge/run_server.sh
       • Windows: /c C:\path\to\claude-lmstudio-bridge\run_server.bat

💻 Usage with Claude

After setting up the bridge tool, you can use the following commands in Claude:

1. Check the connection status of LM Studio:

Can you check if my LM Studio server is running?

2. List available models:

List the available models in my local LM Studio.

3. Generate text using the local model:

Generate a short poem about spring.

4. Send a conversation completion request:

Ask my local LLM: "What is the main function of transformers in machine learning?"

🚑 Troubleshooting

Diagnose LM Studio connection issues

Use the included debugging tool to check your LM Studio connection status:

python debug_lmstudio.py

For more detailed testing:

python debug_lmstudio.py --test-chat --verbose

Common Issues

"Unable to connect to the LM Studio API"

• Ensure that LM Studio is running.
• Verify that the API server in LM Studio is enabled (Settings > API Server).
• Check that the port used (default: 1234) matches the configuration in your .env file.

"No loaded models"

• Open LM Studio and load a model.
• Verify that the model is running correctly.

"MCP package not found"

• Try reinstalling: pip install "mcp[cli]" httpx python-dotenv.
• Ensure you are using Python 3.8 or higher.

"Claude cannot find the bridge tool"

• Check the configuration of Claude Desktop.
• Ensure that the path to run_server.sh or run_server.bat is a correct absolute path.
• Verify that the server script is executable: chmod +x run_server.sh (macOS/Linux).

⚙️ Advanced Configuration

You can customize the behavior of the bridge tool by creating a .env file:

LMSTUDIO_HOST=127.0.0.1
LMSTUDIO_PORT=1234
DEBUG=false

Set DEBUG=true to enable detailed logging for troubleshooting.

📄 License

MIT