LLM quantization severely damages model quality and perplexity

Article URL: https://github.com/ggerganov/llama.cpp/pull/1684

Comments URL: https://news.ycombinator.com/item?id=37951613

Points: 1

# Comments: 0

Read more here: External Link