| .. |
|
composable_kernel @ a9b170b541
|
e2182cc21d
Support page kvcache in AMD ROCm (#1198)
|
3 hónapja |
|
cutlass @ bf9da7b76c
|
5acb532214
Switch to cutlass v3.6.0, fix perf regression for hdim 128 causal
|
3 napja |
|
flash_attn
|
83e41b3ca4
Add custom ops for compatibility with PT Compile (#1139)
|
3 hónapja |
|
flash_attn_ck
|
53a4f34163
Hotfix due to change of upstream api (#1239)
|
3 hónapja |
|
ft_attention
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
10 hónapja |
|
fused_dense_lib
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
10 hónapja |
|
fused_softmax
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
10 hónapja |
|
layer_norm
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
10 hónapja |
|
rotary
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
10 hónapja |
|
xentropy
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
10 hónapja |
Tri Dao
5acb532214
Switch to cutlass v3.6.0, fix perf regression for hdim 128 causal