Search results

[BUG] MLA example is broken for split-kv and larger Q

? - Needs Triage

bugSomething isn't workingSomething isn't working

inactive-30d

inactive-90d

Status: Open.

#2274 In NVIDIA/cutlass;

· divchenko opened on May 1, 2025

[QST] Quantized conv with s8 output and s32 bias

? - Needs Triage

inactive-30d

inactive-90d

questionQuestionQuestion

Status: Open.

#2271 In NVIDIA/cutlass;

· jstoecker opened on Apr 30, 2025

Support N={48, 80, 96, 112, ...} for SM100 EpilogueTileAuto

inactive-30d

inactive-90d

Status: Open (in progress).

NVIDIA/cutlassnumber 2269

#2269 In NVIDIA/cutlass;

· Algy opened on Apr 29, 2025

[BUG] EpilogueTileAuto doesn't work when tile shape N={48, 80, 96, 112, ...} on SM100

? - Needs Triage

bugSomething isn't workingSomething isn't working

inactive-30d

inactive-90d

Status: Open.

#2268 In NVIDIA/cutlass;

· Algy opened on Apr 29, 2025

[QST] Why only N={64, 128, 192, 256} is supported for mxf8f6f4 GeMM?

? - Needs Triage

inactive-30d

inactive-90d

questionQuestionQuestion

Status: Open.

#2260 In NVIDIA/cutlass;

· Algy opened on Apr 24, 2025

[QST]How to write column major output epilogue in cutlass2.x

? - Needs Triage

inactive-30d

inactive-90d

questionQuestionQuestion

Status: Open.

#2258 In NVIDIA/cutlass;

· tlogn opened on Apr 23, 2025

Limit the number of SMs (sm_count) to user-provided value during profiling.

inactive-30d

inactive-90d

Status: Open (in progress).

NVIDIA/cutlassnumber 2257

#2257 In NVIDIA/cutlass;

· manishucsd opened on Apr 22, 2025

[FEA] CVT F32 -> TF32 PTX for sm80

? - Needs Triage

feature requestNew feature or requestNew feature or request

inactive-30d

inactive-90d

Status: Open.

#2254 In NVIDIA/cutlass;

· osayamenja opened on Apr 19, 2025

[FEA][torchinductor-EVT] Allow function source code to be passed directly to EVT tracer

? - Needs Triage

feature requestNew feature or requestNew feature or request

inactive-30d

inactive-90d

Status: Open.

#2244 In NVIDIA/cutlass;

· mlazos opened on Apr 17, 2025

[QST] why use AutoVectoringCopy rather LDSM in hopper mixed gemm?

? - Needs Triage

inactive-30d

inactive-90d

questionQuestionQuestion

Status: Open.

#2239 In NVIDIA/cutlass;

· danielhua23 opened on Apr 13, 2025

[FEA] Blackwell support for python libraries.

? - Needs Triage

feature requestNew feature or requestNew feature or request

inactive-30d

inactive-90d

Status: Open.

#2237 In NVIDIA/cutlass;

· eliphatfs opened on Apr 12, 2025

[BUG] No TMEM allocation in Blackwell CuTe tutorial examples

? - Needs Triage

bugSomething isn't workingSomething isn't working

inactive-30d

inactive-90d

Status: Open.

#2230 In NVIDIA/cutlass;

· allispaul opened on Apr 8, 2025