Search results

[CuTeDSL] Tensor vs TensorSSA, <code>.load()</code> vs <code>cute.copy</code>

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2430 In NVIDIA/cutlass;

· HanGuo97 opened on Jun 29, 2025

[QST]56_hopper_ptr_array_batched_gemm exec error

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2427 In NVIDIA/cutlass;

· youzjuer opened on Jun 27, 2025

[QST] How to implement a mixed-type gemm like fp16/int4 on sm100a (blackwell) with umma?

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2424 In NVIDIA/cutlass;

· JJXiangJiaoJun opened on Jun 26, 2025

[QST]Can I use list of tensor as input in 68_hopper_fp8_warp_specialized_grouped_gemm_with_blockwise_scaling

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2423 In NVIDIA/cutlass;

· youzjuer opened on Jun 26, 2025

[QST] Sychronization primitives in the CuTe DSL

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2418 In NVIDIA/cutlass;

· alexarmbr opened on Jun 24, 2025

[QST] b2b gemm fp32

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2415 In NVIDIA/cutlass;

· rookflying opened on Jun 24, 2025

[QST] Does dense_gemm in blackwell support sm_120?

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2413 In NVIDIA/cutlass;

· liz-badada opened on Jun 24, 2025

[QST]how to use 68_hopper_fp8_warp_specialized_grouped_gemm_with_blockwise_scaling

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2412 In NVIDIA/cutlass;

· youzjuer opened on Jun 23, 2025

When will version 4.0 be released

Status: Open.

#2410 In NVIDIA/cutlass;

· dingjingzhen opened on Jun 22, 2025

[QST] CUTLASS GEMV with very large matrices [50000, 50000]

? - Needs Triage

questionQuestionQuestion

Status: Open.

#2409 In NVIDIA/cutlass;

· qikichen opened on Jun 20, 2025

[BUG] Cutlass and Cute-DSL generate suboptimal code for UMMA that use more registers than necessary

? - Needs Triage

bugSomething isn't workingSomething isn't working

Status: Open.

#2408 In NVIDIA/cutlass;

· tridao opened on Jun 17, 2025

[BUG] Cutlass cannot be compiled with the <code>-G</code> flag for some of the latter examples.

? - Needs Triage

bugSomething isn't workingSomething isn't working

Status: Open.

#2406 In NVIDIA/cutlass;

· mrakgr opened on Jun 16, 2025