Skip to content

Pull requests: NVIDIA/cutlass

New pull request New
103 Open 807 Closed
103 Open 807 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

remove mixed_input_fmha_prefill.py
#3041 opened Feb 18, 2026 by keithzzzzz Loading…
cutlass: enable SM121-gated MXFP4 MoE kernel path
#3038 opened Feb 16, 2026 by christopherowen Loading…
DOC: fix typo in media/docs/cpp/cutlass_3x_design.md
#3035 opened Feb 15, 2026 by kfpanda123 Loading…
[CuTeDSL] Flash Attention v2 for SM120 (Blackwell GeForce)
#3030 opened Feb 13, 2026 by blake-snc Loading…
1
6
Add option to not suffix prints with new line
#3028 opened Feb 13, 2026 by SzymonOzog Loading…
CUTLASS 3x Green Context Support
#3019 opened Feb 11, 2026 by XiaoSongXS Loading…
minor: wrong cordinate in layout algebra docs section
#3014 opened Feb 10, 2026 by JINO-ROHIT Loading…
Declare CUDA standard 20 as requirement for example 63 (fixes #3011)
#3013 opened Feb 10, 2026 by reuterbal Loading…
Fix typo in cute's tutorial example
#3012 opened Feb 10, 2026 by lygztq Loading…
use compiler macro to imporve the compatibility
#3008 opened Feb 6, 2026 by reed-lau Loading…
1
Resolve build warnings in C++20
#2998 opened Feb 3, 2026 by Algy Loading…
add: add comments to help understand
#2993 opened Feb 2, 2026 by meiniangpp416 Loading…
Fix/nvfp4 tensor init
#2989 opened Jan 28, 2026 by michael604work Loading…
Fix mixed_input_fmha_decode example
#2986 opened Jan 26, 2026 by anakinxc Loading…
Fix redundant tile copies in wgmma_sm90 tutorial pipeline loop
#2982 opened Jan 25, 2026 by Johnsonms Loading…
1
Fix error in Blackwell document of referring to Mxf4 format as NVF4
#2977 opened Jan 23, 2026 by zianglih Loading…
2
fix(examples): fix device compatibility check for Ada FP8 GEMM inactive-30d
#2954 opened Jan 13, 2026 by w1ndseeker Loading…
1
Update profiler.md with how to use generator.py inactive-30d
#2943 opened Jan 10, 2026 by aidando73 Loading…
1
cutlass profiler - align emitted SFA/SFB kernel naming with typical convention inactive-30d
#2942 opened Jan 10, 2026 by aidando73 Loading…
1
Fix Warp Memory Access Arrangement in Epilogue: Upper Bound memory access width by output tile width inactive-30d
#2938 opened Jan 8, 2026 by lukas-ruettgers Loading…
1
1
add missing Mma-specialization for m16n8k32.s32.s4.s4.s32 inactive-30d
#2936 opened Jan 8, 2026 by dxqb • Draft
1
feat(examples/test_run): use runtime sm arch
#2916 opened Dec 31, 2025 by tpoisonooo Loading…
6
Fix finding cuDNN inactive-30d
#2890 opened Dec 19, 2025 by TLescoatTFX Loading…
1
4
docs: Add FP16 GEMM documentation to sgemm_sm80.cu - Fixes #1686 inactive-30d
#2870 opened Dec 10, 2025 by blueberrycongee Loading…
1
2
[WIP]Unit tests for Kernels that perform BF16 x BF16 = MXFP8 and MXFP8 x MXFP8 = BF16 inactive-30d
#2857 opened Dec 8, 2025 by Shreya-gaur Loading…
1
ProTip! no:milestone will show everything without a milestone.