Skip to content

Pull requests: NVIDIA/TransformerEngine

New pull request New
166 Open 2,400 Closed
166 Open 2,400 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

[PyTorch] Decline fused grouped MLP when the backward format is not E4M3 community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3352 opened Aug 12, 2026 by wilyan09007 Loading…
5 of 13 tasks
1
1
[PyTorch] Fused GDN attention 2.19
#3351 opened Aug 12, 2026 by ksivaman Member • Draft
9 of 15 tasks
2
[Pytorch] MOE Sequential Block
#3350 opened Aug 11, 2026 by vthumbe1503 Collaborator • Draft
13 tasks
[Common] Upgrade nccl-extensions commit
#3349 opened Aug 11, 2026 by phu0ngng Collaborator Loading…
13 tasks
2
[PyTorch] Fix MXFP8 master weight cast when a rank owns no shard. community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3348 opened Aug 11, 2026 by rapatel Loading…
8 of 13 tasks
3
[PyTorch] Make the Triton mask-map permute respect num_out_tokens community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3347 opened Aug 11, 2026 by truong-v Loading…
2 tasks done
1
@tdophung
1
[Pytorch] Fix swizzle scale output shape in variable shape case
#3344 opened Aug 11, 2026 by YangFei1990 Collaborator Loading…
8 of 13 tasks
@vthumbe1503
4
[PyTorch] Enable mid-training recipe update
#3343 opened Aug 11, 2026 by negvet Collaborator • Draft
1 of 13 tasks
Stabilize FlashAttention CI environments
#3341 opened Aug 11, 2026 by sudhakarsingh27 Member Loading…
@sudhakarsingh27
1
Use a plain parameter for high-precision grouped linear weights community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3340 opened Aug 10, 2026 by wujingyue Contributor • Draft
Optionally pad QK/V for better kernel selection community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3339 opened Aug 10, 2026 by janEbert Loading…
5 of 13 tasks
@cyanguwa
11
[Common] Improved performance of Group MXFP8 kernels 2.19
#3337 opened Aug 10, 2026 by Oleg-Goncharov Collaborator Loading…
6 of 13 tasks
@ptrendx
3
[PyTorch] Fine-grained recipe docs documentation Improvements or additions to documentation
#3336 opened Aug 10, 2026 by negvet Collaborator Loading…
1 of 13 tasks
@pggPL
2
[PyTorch] Add torch_stable compatibility layer for incremental migration to the torch stable ABI
#3335 opened Aug 10, 2026 by pggPL Collaborator • Draft
7 of 13 tasks
Add a backwards linear function to be used with the fused mla q up-proj 2.19 community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3330 opened Aug 7, 2026 by chaseblock Contributor Loading…
13 tasks
@cyanguwa
4
[PyTorch] Fix deferred initialization in fusible ops
#3327 opened Aug 7, 2026 by denera Collaborator Loading…
8 of 13 tasks
1
@denera
9
Refactor GroupedLinear quantization dispatch
#3326 opened Aug 7, 2026 by negvet Collaborator Loading…
1 of 13 tasks
@vthumbe1503
10
Prototype NVFP4 with FP8 UE5M3 block scales 2.19 enhancement New feature or request
#3325 opened Aug 7, 2026 by timmoon10 Member • Draft
5 of 13 tasks
@ptrendx
1
[PyTorch] Enable NVFP4 row-scaled (per-token) backward for GroupedLinear community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3324 opened Aug 7, 2026 by cael-ling Contributor Loading…
1 of 13 tasks
@ptrendx
5
[Pytorch] Enable TE Op to consume extra_outputs from a previously run Op in TE Sequential 2.19
#3320 opened Aug 5, 2026 by vthumbe1503 Collaborator Loading…
13 tasks
@ptrendx
33
[Common][PyTorch] Fuse the RHT into grouped NVFP4 quantize on non-SM100 architectures community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3317 opened Aug 5, 2026 by davidkny22 Contributor Loading…
6 of 13 tasks
@Oleg-Goncharov
3
[Common/PyTorch] Grouped weighted-SwiGLU MXFP8 kernel community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3315 opened Aug 4, 2026 by cael-ling Contributor Loading…
3 of 13 tasks
@Oleg-Goncharov
12
Log when thd with dropout falls to the composite cuDNN engine community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3313 opened Aug 4, 2026 by bzantium Loading…
@KshitijLakhani
8
[CI] Publish GB200 aarch64 wheel artifacts community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3311 opened Aug 4, 2026 by bvolpato Loading…
5 of 13 tasks
@fheinecke
3
nvrtc MXFP8 kernels community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#3302 opened Aug 3, 2026 by CarlosGomes98 Contributor • Draft
13 tasks
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.