"author:zbrdc" matching MCP servers:

Routes AI tasks to appropriate local LLM models (quick, coder, MoE, thinking) with automatic model selection, multi-backend support (Ollama, llama.cpp, Gemini), and parallel processing capabilities.