Production System
72,400 GitHub stars
llama.cpp
Port of Facebook's LLaMA model in pure C/C++ without external dependencies. Powers on-device LLM inference on MacBook Metal, Android, and CPU with 4-bit GGUF quantization.
C++MetalGGUFQuantizationEdge AI
Open Source • C++ • Metal • GGUF • Quantization • Edge AI
Interactive Quickstart Guide:
# 1. Clone repository git clone https://github.com/ggerganov/llama.cpp.git cd ggerganov-llama-cpp # 2. Install dependencies with npm npm install # 3. Configure environment variables cp .env.example .env.local # 4. Launch local dev server npm run dev
