david/flash-attention

spogulis no https://github.com/Dao-AILab/flash-attention

Autors	SHA1 Ziņojums	Datums
Jay Shah	483b26e541 add working fp8 varlen	2 mēneši atpakaļ
Son Nguyen	478ee666cc Make namespace comment consistent (#1305)	2 mēneši atpakaļ
jayhshah	a5a75274bc FA3 kvcache + split kv + gqa parallelization (#1236)	2 mēneši atpakaļ
Ying Zhang	dfe1a59e4b Add var-seq-len to FA3 fp16 / bf16 fwd (#1072)	5 mēneši atpakaļ