Skip to content

Pull requests: NVIDIA/TransformerEngine

New pull request New
126 Open 1,880 Closed
126 Open 1,880 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 deprecated qkv/kv_packed apis
#2696 opened Feb 20, 2026 by sudhakarsingh27 Loading…
13 tasks
2
[PyTorch] Zero-initialize learnable softmax_offset in DotProductAttention
#2694 opened Feb 20, 2026 by fjosw Loading…
7 of 13 tasks
2
Enable sm120 support for fused attn if cuDNN is 9.18.1+
#2693 opened Feb 20, 2026 by KshitijLakhani • Draft
13 tasks
[JAX] Fix batcher in FusedAttn primitive for when seg ids bdims != seg pos bdims 2.13.0
#2692 opened Feb 19, 2026 by KshitijLakhani Loading…
6 of 13 tasks
1
@KshitijLakhani
7
NVFP4 primary weight support
#2691 opened Feb 19, 2026 by WanZzzzzz Loading…
10 of 13 tasks
41
[PyTorch] Error out if constructing LayerNormLinear with row tensor parallelism bug Something isn't working
#2688 opened Feb 17, 2026 by timmoon10 Loading…
6 of 13 tasks
7
[PyTorch] torch.compile support for permutation functions
#2686 opened Feb 17, 2026 by pggPL Loading…
9 of 13 tasks
6
[Common] Remove volatile keyword in fused router kernel utils
#2683 opened Feb 13, 2026 by denera • Draft
8 of 13 tasks
@denera
[JAX] Integrate BF16 Grouped GEMM with on-device group sizes
#2680 opened Feb 13, 2026 by jberchtold-nvidia • Draft
8 of 13 tasks
1
10
[PyTorch][Fused Attn] Add support for cuDNN to return Softmax Stats always and Max when return_max_logit=True
#2677 opened Feb 12, 2026 by sudhakarsingh27 Loading…
7 of 13 tasks
26
[PyTorch] Add dtype information to QuantizedTensorStorage class
#2676 opened Feb 12, 2026 by ptrendx Loading…
1 of 13 tasks
13
[Common] MOE Split dBias cpu_overhead enhancement New feature or request MoE
#2674 opened Feb 11, 2026 by Oleg-Goncharov Loading…
8 of 13 tasks
15
Cherry pick "Adds dst.dtype information in copy_ method of quantized tensors. #2120"
#2673 opened Feb 11, 2026 by ptrendx Loading…
13 tasks
4
Fix and harden attention benchmark script
#2671 opened Feb 10, 2026 by KshitijLakhani • Draft
13 tasks
@KshitijLakhani
Fix nvfp4 convert_and_update_tensor shape check
#2670 opened Feb 10, 2026 by skydoorkai Loading…
13 tasks
1
13
Pytorch binding for cublas gemm + Grouped Linear integration
#2669 opened Feb 10, 2026 by vthumbe1503 • Draft
13 tasks
[pyTorch] Replace the make_empty implementation to use C++ implementation
#2666 opened Feb 10, 2026 by ptrendx Loading…
1 of 13 tasks
6
Improve handling of the NVTE_CUDA_ARCHS
#2665 opened Feb 9, 2026 by ptrendx Loading…
1 of 13 tasks
14
Add multi-precision training support to FSDP script
#2662 opened Feb 9, 2026 by aagallo Loading…
6 of 13 tasks
26
[JAX] Fix FSDP when FSDP+EP is active
#2649 opened Feb 3, 2026 by jberchtold-nvidia Loading…
8 of 13 tasks
6
start devcontainer for transformerEngine
#2646 opened Feb 3, 2026 by pstjohn • Draft
[PyTorch Debug] Support tensor dump
#2645 opened Feb 3, 2026 by pggPL • Draft
13 tasks
Add NVTE_KEEP_BACKWARD_UNQUANTIZED
#2644 opened Feb 3, 2026 by zianglih • Draft
6 of 13 tasks
101
ci(fix): NGC build
#2643 opened Feb 2, 2026 by ko3n1g Loading…
13 tasks
13
Add examples for MoE models - Mixtral in TE
#2642 opened Feb 2, 2026 by faradawn Loading…
2 of 13 tasks
1
5
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.