Skip to content

Run the vector SDPA kernels at head_dim 256 on CUDA - #4698

Merged
zcbenz merged 1 commit into
ml-explore:mainfrom
yanghaojin:gbx/sdpa-vector-hd256
Oct 16, 2026
Merged

zcbenz merged 1 commit into
ml-explore:mainfrom
yanghaojin:gbx/sdpa-vector-hd256