VPTQ: Extreme Low-bit Vector Post-Training Quantization for Large Language Models Paper • 2409.17066 • Published Sep 25, 2024 • 28
