LM Studio

🚀 LMStudio-MCP Project Documentation

LMStudio-MCP creates a bridge that connects Claude (with MCP functionality) to your locally running LM Studio instance. This enables Claude to:

• Check the health status of the LM Studio API.
• List available models.
• Get the currently loaded model.
• Generate completions using your local models.

Through this bridge, you can combine Claude's capabilities with your own private models.

✨ Features

The bridge provides the following features:

• health_check(): Verify if the LM Studio API is accessible.
• list_models(): Get a list of all available models in LM Studio.
• get_current_model(): Identify the currently loaded model.
• chat_completion(prompt, system_prompt, temperature, max_tokens): Generate text from your local model.

📦 Installation

Prerequisites

• Python 3.7 or higher.
• LM Studio installed and running (https://lmstudio.ai/).
• A model loaded into LM Studio.
• Claude with MCP access.
• Required Python packages (see installation steps).

Steps

1. Clone this repository:

   git clone https://github.com/infinitimeless/LMStudio-MCP.git
   cd LMStudio-MCP
   

2. Install the required packages:

   pip install requests "mcp[cli]" openai
   

📚 Configuration

MCP Configuration

To allow Claude to connect to this bridge, you need to configure the MCP settings correctly. You can choose:

1. Use directly from GitHub:

   {
     "lmstudio-mcp": {
       "command": "uvx",
       "args": [
         "https://github.com/infinitimeless/LMStudio-MCP"
       ]
     }
   }
   

2. Use local installation:

   {
     "lmstudio-mcp": {
       "command": "/bin/bash",
       "args": [
         "-c",
         "cd /path/to/LMStudio-MCP && source venv/bin/activate && python lmstudio_bridge.py"
       ]
     }
   }
   

For detailed instructions on MCP configuration, see MCP_CONFIGURATION.md.

💻 Usage Examples

Basic Usage

1. Start your LM Studio application and ensure it is running on the default port 1234.

2. Load a model in LM Studio.

3. If running locally (without using uvx), start the LMStudio-MCP server:

   python lmstudio_bridge.py
   

4. In Claude, select "lmstudio-mcp" to connect to the MCP server.

🔧 Technical Details

Known Limitations

• Some models (e.g., phi-3.5-mini-instruct_uncensored) may have compatibility issues.
• The bridge currently only uses the OpenAI-compatible API endpoint of LM Studio.
• The model's responses will be limited by the capabilities of your locally loaded model.

Troubleshooting

API Connection Issues

If Claude reports a 404 error and cannot connect to LM Studio:

• Ensure that LM Studio is running and a model is loaded.
• Check if the LM Studio server is running on port 1234.
• Verify that the firewall is not blocking the connection.
• If the problem persists, try using "127.0.0.1" instead of "localhost" in the API URL.

Model Compatibility

If some models do not work properly:

• Ensure that the model you are using is compatible with LM Studio.
• Refer to MCP_CONFIGURATION.md for more information.

For more troubleshooting steps, see TROUBLESHOOTING.md.

📄 License

Notes

• This project is still in development, and there may be instability or incomplete functionality.
• Please read the relevant documentation carefully and back up important data before use.