llama.cpp

github.com/ggerganov/llama.cpp

C/C++ implementation for running quantized LLMs on consumer hardware (CPU, GPU, Apple Silicon). — extracted from the official website or Wikipedia.