Skip to content

Tune FP8FlashAttentionSm120Tma: N=32 default, BF16 output mode

a3402a1
Select commit
Loading
Failed to load commit list.
Open

[CuTeDSL] Flash Attention v2 for SM120 (Blackwell GeForce) #3030

Tune FP8FlashAttentionSm120Tma: N=32 default, BF16 output mode
a3402a1
Select commit
Loading
Failed to load commit list.

Workflow runs completed with no jobs