EfficientQAT: LLM Quantization, gets a 2-bit llama2-70B outperform regular 13B
Article URL: https://old.reddit.com/r/LocalLLaMA/comments/1e5x2k4/new_llms_quantization_algorithm_efficientqat/
Comments URL: https://news.ycombinator.com/item?id=40991588
Points: 7
# Comments: 0
Read more here: External Link