I stress-tested LLM quantization by deliberately breaking models (github.com) 1 points by GraceJackson07 1mo ago ↗ HN
2 comments
[ 3.2 ms ] story [ 19.4 ms ] thread