Llama.cpp Bridge
Connects Claude Desktop to your local llama.cpp models, letting you chat with local LLMs directly through Claude's interface.
82,364 views5Local (stdio)
What it does
- Chat with local llama.cpp models through Claude Desktop
- Control generation parameters like temperature and max_tokens
- Monitor llama-server health and status
- Track performance metrics and token usage
- Test model capabilities with built-in tools
Best for
AI researchers running local modelsPrivacy-focused users avoiding cloud APIsDevelopers integrating local LLMs with desktop workflows
No cloud API keys requiredFull conversation supportBuilt-in testing tools
