Llama.cpp Bridge
Bridges local llama-server instances with MCP clients, providing chat interface, health monitoring, and configurable generation pa
This MCP server provides a bridge between Claude Desktop and local LibreModel (Gigi) instances running via llama-server, built by Jerry using TypeScript with the Model Context Protocol SDK and Zod validation. The implementation offers three core tools: a main chat interface with full parameter control (temperature, max tokens, top-p, top-k sampling), quick predefined tests for capability verification, and health checks with server diagnostics including props endpoint querying. Features configurable defaults, graceful error handling with timeout protection, and comprehensive usage documentation through built-in resources, making it valuable for users who want to integrate their local llama.cpp models with Claude Desktop while maintaining full control over generation parameters and server connectivity.
Source
Repository: https://github.com/openconstruct/llama-mcp-server
Maintain Llama.cpp Bridge?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[Llama.cpp Bridge on getagentictools](https://getagentictools.com/mcp/openconstruct-llama-mcp-server?ref=badge) Related mcp
CLI2MCP Wraps any CLI binary as an MCP server by parsing its --help output and synthesizing a JSON Schema at startup.
Zerocracy Provides a bridge between Zerocracy's project management platform and development teams, offering actionable recommendations to im
CC2WX WeChat Bridge Bridge between Claude Code and WeChat via iLink Bot API with permission relay.
Apache Unomi Maintain persistent user profiles and context across conversations.
GitHub Pages Documentation Automates GitHub Pages documentation deployment by analyzing repositories, recommending static site generators like Jekyll, Hugo,
WeChat Mini-Program Agent Automate and test WeChat Mini Programs via DevTools with navigation, screenshots, element interaction, and DOM inspection.