david/flash-attention

Mirror von https://github.com/Dao-AILab/flash-attention

Autor	SHA1 Nachricht	Datum
jayhshah	a5a75274bc FA3 kvcache + split kv + gqa parallelization (#1236)	vor 2 Monaten
jayhshah	c92ca63268 FA3 FP8 qkv descales + restore max offset for h128 causal + added sync for producer WG (#1173)	vor 3 Monaten
jayhshah	5018ac6ac5 Fp8 kernel with "in-kernel" transpose of V in producer (#1100)	vor 4 Monaten
Tri Dao	3aae9c18c1 Revert "Changes For FP8 (#1075)"	vor 4 Monaten
ganeshcolfax	1899c970c8 Changes For FP8 (#1075)	vor 4 Monaten