Commit Verlauf

Autor SHA1 Nachricht Datum
  Jay Shah 50cb90aea6 comment out unimplemented kwargs from flash_attn_with_kvcache vor 4 Monaten
  Jay Shah b5cac6d586 rebase with Is_local disabled temporarily vor 5 Monaten
  Jay Shah 2472e5e0b4 add 'in principle' fp8 kv cache support vor 5 Monaten
  Jay Shah e36e004cb3 change fp8 code path to allow for split kernel and kv cache without perf regression vor 5 Monaten
  Jay Shah f53703b98d add split kv heuristic modifications vor 5 Monaten
  Ganesh Bikshandi 099ca28488 add gqa decoding logic. vor 5 Monaten
  Jay Shah a06f1f970b add to kv cache api vor 6 Monaten
  Jay Shah 535b8279bb complete gqa parallel changes for non-causal vor 6 Monaten
  Ganesh Bikshandi 84e31c22d6 adding python interface. vor 6 Monaten
  Ying Zhang dff976a84a fixes vor 6 Monaten
  Ying Zhang 7b4e68e04f hopper local attention vor 6 Monaten
  Ying Zhang db80387343 Add seqused_q in fwd / bwd and seqused_k in bwd. vor 6 Monaten
  Charlene Yang bdf733be55 Add q, k, v descales to FA3 interface (#1210) vor 5 Monaten
  jayhshah c92ca63268 FA3 FP8 qkv descales + restore max offset for h128 causal + added sync for producer WG (#1173) vor 6 Monaten
  Tri Dao bafe253042 [FA3] Bwd vor 7 Monaten
  Tri Dao 3aae9c18c1 Revert "Changes For FP8 (#1075)" vor 7 Monaten
  ganeshcolfax 1899c970c8 Changes For FP8 (#1075) vor 7 Monaten
  Ying Zhang dfe1a59e4b Add var-seq-len to FA3 fp16 / bf16 fwd (#1072) vor 7 Monaten
  youkaichao ef3e358a25 remove lambda (#1056) vor 7 Monaten
  Tri Dao 7f67966cc7 FA3 initial code release vor 7 Monaten