Skip to content

[PyTorch] Enable fused FP8 block-scaling path in GroupedLinear module#3171

Open
denera wants to merge 10 commits into
NVIDIA:mainfrom
denera:pytorch/grouped-linear-fused-fp8bs
Open

[PyTorch] Enable fused FP8 block-scaling path in GroupedLinear module#3171
denera wants to merge 10 commits into
NVIDIA:mainfrom
denera:pytorch/grouped-linear-fused-fp8bs

Commits

Commits on Jul 9, 2026