Skip to content

Single-launch CUTLASS grouped GEMM for per-tensor NVFP4#3134

Open
cael-ling wants to merge 2 commits into
NVIDIA:mainfrom
cael-ling:optimize/group-gemm
Open

Single-launch CUTLASS grouped GEMM for per-tensor NVFP4#3134
cael-ling wants to merge 2 commits into
NVIDIA:mainfrom
cael-ling:optimize/group-gemm

Commits

Commits on Jun 17, 2026