NVIDIA/cutlass v4.2.2

CUTLASS 4.2.2

on GitHub

CUTLASS C++

• Make version.h NVRTC JIT compilation compatible.
• Allow linking large cutlass library on 64bit platform.
• Fix alignment-related miscalculation for pipeline stages of Blackwell blockscaled GEMM.
• Fix for blockwise group gemm nosmem epilogues and no sfd with nosmem group gemm epilogues.