Search results

[QST] [CuTeDSL] sm100 get_smem_store_op layout mismatch between tcgen05.ld 16x256b and stmatrix m8n8

? - Needs Triage

questionQuestionQuestion

Status: Open.

#3099 In NVIDIA/cutlass;

· Edenzzzz opened on Mar 9, 2026

SM120 (Bug) (With FIx)(RTX Blackwell) NVFP4 MoE: CUTLASS Grouped GEMM Produces Garbage Output; Fixed via FlashInfer SM120 Patches + compute_120f (CUDA 13.0) — 39 tok/s Native FP4

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3096 In NVIDIA/cutlass;

· brandonmmusic-max opened on Mar 9, 2026

[FEA] TVM FFI Support for Dataclass

? - Needs Triage

CuTe DSL

feature requestNew feature or requestNew feature or request

Status: Open.

#3094 In NVIDIA/cutlass;

· HanGuo97 opened on Mar 7, 2026

[BUG] tcgen05 ld.red CSEd incorrectly

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3090 In NVIDIA/cutlass;

· skrider opened on Mar 6, 2026

[QST] Wrong comment of LayoutType::B128_BASE32B in make_umma_desc?

? - Needs Triage

questionQuestionQuestion

Status: Open.

#3089 In NVIDIA/cutlass;

· bluebabyxp opened on Mar 6, 2026

[DOC] [CuTe DSL] Wrong output shape for logical_product?

? - Needs Triage

documentationDocumentationDocumentation

Status: Open.

#3085 In NVIDIA/cutlass;

· Edenzzzz opened on Mar 1, 2026

[BUG] <code>import cutlass</code> crashes on CPU-only machines when jax is installed (v4.4 regression)

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3083 In NVIDIA/cutlass;

· henrylhtsang opened on Feb 27, 2026

[FEA] will cuteDSL support INT8 GEMM and FP8 GEMM on Ada/Ampere platform

? - Needs Triage

CuTe DSL

feature requestNew feature or requestNew feature or request

Status: Open.

#3081 In NVIDIA/cutlass;

· houqi opened on Feb 27, 2026

[QST] TiledMMA.make_fragment_A fails with non-swizzled smem layout

? - Needs Triage

questionQuestionQuestion

Status: Open.

#3078 In NVIDIA/cutlass;

· HarryWu99 opened on Feb 27, 2026

[BUG] TiledMMA modifier can't be set in different jit function

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3077 In NVIDIA/cutlass;

· HarryWu99 opened on Feb 27, 2026

[DOC] Storing 2D Tile From Accumulator Registers To Global Memory Using TMA.

? - Needs Triage

documentationDocumentationDocumentation

Status: Open.

#3076 In NVIDIA/cutlass;

· tugrul512bit opened on Feb 26, 2026

[QST] Is _cast used anywhere in CuTeDSL ?

? - Needs Triage

questionQuestionQuestion

Status: Open.

#3068 In NVIDIA/cutlass;

· Peter9606 opened on Feb 25, 2026