Tri Dao
|
180ff782dd
Template for Sm86
|
1 viikko sitten |
Tri Dao
|
7bc3f031a4
Compile for both Sm80 and Sm90
|
1 viikko sitten |
Tri Dao
|
7a802796e1
Big refactor and update
|
1 viikko sitten |
Kai Londenberg
|
b443207c1f
Paged Attention support for FA3 (#1268)
|
2 kuukautta sitten |
jayhshah
|
a5a75274bc
FA3 kvcache + split kv + gqa parallelization (#1236)
|
3 kuukautta sitten |
Ying Zhang
|
dfe1a59e4b
Add var-seq-len to FA3 fp16 / bf16 fwd (#1072)
|
5 kuukautta sitten |
Tri Dao
|
7f67966cc7
FA3 initial code release
|
6 kuukautta sitten |