ModelCloud / GPTQModel

git clone https://github.com/ModelCloud/GPTQModel.git

LLM model quantization (compression) toolkit with HW acceleration support for Nvidia, AMD, Intel GPU and Intel/AMD/Apple CPU via HF, vLLM, and SGLang.