Skip to content

Pull requests: NVIDIA/TransformerEngine

New pull request New
104 Open 1,804 Closed
104 Open 1,804 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

ONNX: Fix FP8 quantization for the second MLP in LayerNormMLP
#2577 opened Jan 8, 2026 by victoroliv2 Loading…
1 of 13 tasks
1
Debug doc generation bug Something isn't working documentation Improvements or additions to documentation
#2576 opened Jan 7, 2026 by timmoon10 Loading…
9 of 13 tasks
3
fix(examples): te_llama compatibility with transformers >= 4.57
#2572 opened Jan 7, 2026 by sbhavani Loading…
6 of 13 tasks
1
5
Update THD sink attention logic for cudnn >=9.18.0
#2568 opened Jan 6, 2026 by cuichenx Loading…
13 tasks
5
CPU Optimizations for FP8 cpu_overhead
#2559 opened Jan 5, 2026 by vthumbe1503 Loading…
13 tasks
7
[NVFP4][Dense/MoE] Integrate Cutlass NVFP4 Row-Cast-Col-RHT-Transpose-Cast Fusion Kernel fp4 MoE
#2555 opened Jan 3, 2026 by zhongbozhu Loading…
2 of 16 tasks
@zhongbozhu
3
[Pytorch] Enhance bf16 precision optimizer performance with memory buffer
#2551 opened Dec 31, 2025 by Baidu-AIAK Loading…
4
[PyTorch] Remove unnecessary save of weights
#2549 opened Dec 30, 2025 by pggPL • Draft
8 of 13 tasks
[PyTorch]Add Casting-Free FP8-Flow-MoE Blockwise Optimizations community-contribution PRs from external contributor outside the core maintainers, representing community-driven work.
#2544 opened Dec 26, 2025 by xiaoxi-wangfj Loading…
4 of 13 tasks
9
[JAX] Refactor and trim TE JAX Attn testing
#2542 opened Dec 24, 2025 by KshitijLakhani • Draft
13 tasks
@KshitijLakhani
[PyT] Plumbing correct bias dims from TE to cudnn attention bug Something isn't working pytorch
#2537 opened Dec 20, 2025 by KshitijLakhani Loading…
4 of 11 tasks
@KshitijLakhani
4
[PyTorch] Bunch of fixes for cpu offloading
#2535 opened Dec 19, 2025 by pggPL • Draft
13 tasks
1
[DO NOT MERGE] Get seqlens and offsets in O(N) space instead of O(N*N) space do not merge
#2530 opened Dec 17, 2025 by KshitijLakhani • Draft
13 tasks
@KshitijLakhani
[JAX] Calculate seqlens and offsets in O(N) space instead of O(N*N) space for THD sequences attention
#2522 opened Dec 16, 2025 by KshitijLakhani • Draft
13 tasks
@KshitijLakhani
Enable post-RHT amax estimation with separate amax scale kernel fp4
#2521 opened Dec 16, 2025 by negvet • Draft
13 tasks
2
Documentation for cpu offloading documentation Improvements or additions to documentation
#2520 opened Dec 16, 2025 by pggPL Loading…
8 of 13 tasks
2
[JAX] HLO FFI tests jax
#2517 opened Dec 16, 2025 by jberchtold-nvidia Loading…
7 of 13 tasks
11
Cpu optimizations v2 cpu_overhead
#2514 opened Dec 12, 2025 by vthumbe1503 • Draft
13 tasks
[DO NOT MERGE] Testing v2.6 + pr2201 attention
#2513 opened Dec 12, 2025 by KshitijLakhani • Draft
13 tasks
@KshitijLakhani
[Common] Optimize fused RoPE kernel performance performance Performance issues
#2508 opened Dec 11, 2025 by yaox12 • Draft
13 tasks
[common] Add support for cuBLASLt GEMM for GroupedTensor MoE
#2502 opened Dec 10, 2025 by pggPL Loading…
8 tasks done
1
44
Add logic for block-scaled tensors with GEMM swizzled scales enhancement New feature or request MoE performance Performance issues refactor
#2486 opened Dec 6, 2025 by timmoon10 Loading…
14 of 19 tasks
1
19
[JAX] Estimate post-RHT amax using regular amax fp4
#2479 opened Dec 4, 2025 by jberchtold-nvidia • Draft
13 tasks
2
Add support for SWA (left, right) with FusedAttention 2.12.0
#2477 opened Dec 4, 2025 by sudhakarsingh27 Loading…
22 of 28 tasks
35
[JAX] Einsum with quantization
#2474 opened Dec 3, 2025 by phu0ngng • Draft
13 tasks
ProTip! Updated in the last three days: updated:>2026-01-04.