FlashQwen – A from-scratch CUDA inference engine for Qwen3
github.com5 pointsby langtang19960 comments
github.com · 137 points · 285 comments
github.com · 13 points · 3 comments
github.com · 2 points · 0 comments
forgottenbytes.net · 227 points · 78 comments
fata.dev · 109 points · 51 comments
machine0.io · 90 points · 35 comments