Driss Guessous bc482cbf91 Add a macro for namespace (#1419) 3 settimane fa
..
composable_kernel @ 888317e698 77ad12d24e [AMD ROCm] Support variable length of page attention (#1431) 4 settimane fa
cutlass @ c506e16788 68bf390920 Update Cutlass to fix mem fence 1 mese fa
flash_attn bc482cbf91 Add a macro for namespace (#1419) 3 settimane fa
flash_attn_ck 74aed78373 Replace c10::optional with std::optional in flash_attn 3 settimane fa
ft_attention 74aed78373 Replace c10::optional with std::optional in flash_attn 3 settimane fa
fused_dense_lib 74aed78373 Replace c10::optional with std::optional in flash_attn 3 settimane fa
fused_softmax 50896ec574 Make nvcc threads configurable via environment variable (#885) 11 mesi fa
layer_norm 74aed78373 Replace c10::optional with std::optional in flash_attn 3 settimane fa
rotary 1feb711f46 Fix compilation with clang on ARM64 (#1285) 2 mesi fa
xentropy 1feb711f46 Fix compilation with clang on ARM64 (#1285) 2 mesi fa