Story Detail of id 50222723 | Liveview Hacker News

Hacker News new | past | comments | ask | show | jobs | submit

xenonite18 hours ago | on: GPU-unlock: FP/BF16 instruction issue rate on Nvidia 3000's cards

great to hear, which quant + KV quant + KV size are you on? is it llama.cpp?

aunty_helen7 hours ago | parent

Here's the writeup from the experiments that claude did: https://claude.ai/artifact/X6WJVtKdL1LrhYuCMqdXvG#b9a4d3e2-8...