Super HN

New Show
   NanoQuant: Efficient Sub-1-Bit Quantization of Large Language Models (arxiv.org)