Article URL: https://www.bbc.com/news/articles/c4gmggp2y99o
Comments URL: https://news.ycombinator.com/item?id=43517833
Points: 74
# Comments: 155
Vytvořeno
2mo
|
30. 3. 2025 2:30:19
Chcete-li přidat komentář, přihlaste se
Ostatní příspěvky v této skupině

I discovered that in LLM inference, keys and values in the KV cache have very different quantization sensitivities. Keys need higher precision than values to maintain quality.
I patched llama.cp
Article URL: https://clojurescript.org/news/2025-05-16-release




Article URL: https://bobacollection.staxmuseum.org/
Comments URL: https://news.y