.. |
composable_kernel @ 13332998a4
|
88d1657a14
[AMD ROCm] Fix KVcache bug and improve performance (#1328)
|
1 hónapja |
cutlass @ 4c42f73fda
|
61a23ea8a2
Update to Cutlass 3.6.0
|
1 hete |
flash_attn
|
61a23ea8a2
Update to Cutlass 3.6.0
|
1 hete |
flash_attn_ck
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 hete |
ft_attention
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 hete |
fused_dense_lib
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 hete |
fused_softmax
|
50896ec574
Make nvcc threads configurable via environment variable (#885)
|
9 hónapja |
layer_norm
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 hete |
rotary
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 hete |
xentropy
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 hete |