Tri Dao 26a6e0a048 Switch to Cutlass 3.6.0 1 ماه پیش
..
composable_kernel @ a9b170b541 e2182cc21d Support page kvcache in AMD ROCm (#1198) 3 ماه پیش
cutlass @ 19f51596e8 26a6e0a048 Switch to Cutlass 3.6.0 1 ماه پیش
flash_attn 83e41b3ca4 Add custom ops for compatibility with PT Compile (#1139) 2 ماه پیش
flash_attn_ck 53a4f34163 Hotfix due to change of upstream api (#1239) 2 ماه پیش
ft_attention 50896ec574 Make nvcc threads configurable via environment variable (#885) 9 ماه پیش
fused_dense_lib 50896ec574 Make nvcc threads configurable via environment variable (#885) 9 ماه پیش
fused_softmax 50896ec574 Make nvcc threads configurable via environment variable (#885) 9 ماه پیش
layer_norm 50896ec574 Make nvcc threads configurable via environment variable (#885) 9 ماه پیش
rotary 50896ec574 Make nvcc threads configurable via environment variable (#885) 9 ماه پیش
xentropy 50896ec574 Make nvcc threads configurable via environment variable (#885) 9 ماه پیش