NVIDIA/cutlass v4.3.1

CUTLASS 4.3.1

Repository: NVIDIA/cutlass

Tag: v4.3.1

Published: 2025-12-02T03:22:01Z

Prerelease: no

Release notes:

CuTe DSL

• New features
• Added Blackwell SM103 support
• Multiple dependent DSOs in the wheel have been merged into one single DSO
• Bug fixing and improvements
• Fixed device reset issue with tvm-ffi
• Fixed tvm-ffi export compiled function

CUTLASS C++

• Support blockscaled variant of ragged contiguous grouped gemm with the new simplified MoE API in example 92.
• The new example works for all microscaling types.