NVIDIA/cutlass v4.3.4

CUTLASS 4.3.4

Repository: NVIDIA/cutlass

Tag: v4.3.4

Published: 2025-12-24T05:49:23Z

Prerelease: no

Release notes:

CuTe DSL

• New features
• Added PDL support along with example Kernel launch with Programmatic Dependent Launch

• Bug fixing and improvements
• Fixed a frame refcnt issue with cuda graph
• Enhancement for tvm-ffi AoT case for earlier module unload
• Fixed order issue in make_smem_layout_a in utils/hopper_helpers.py

CUTLASS C++

• Work around a driver TMA descriptor related bug which will cause occasionally errors on Blackwell when the tensor's backing memory allocation is less than 128KB and it is not a dense non-overlapping tensor.