Skip to content

Pull requests: pggPL/TransformerEngine

New pull request New
32 Open 16 Closed
32 Open 16 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

[CI] Make GitHub Build workflow fit the hosted runner (test PR)
#49 opened Sep 8, 2026 by pggPL Owner Loading…
[PyTorch] Docs: fine-grained quantization recipes (rework of NVIDIA#3336)
#48 opened Sep 7, 2026 by pggPL Owner Loading…
5 of 13 tasks
[PyTorch][torch.compile] Differentiable fp8_output and fp8_grad for Linear
#47 opened Sep 4, 2026 by pggPL Owner Loading…
8 of 13 tasks
[PyTorch] torch.compile support for FusedAttention (diff vs #3480 only)
#46 opened Sep 4, 2026 by pggPL Owner Loading…
[PyTorch] torch.compile for an OperationFuser group holding one operation
#45 opened Sep 4, 2026 by pggPL Owner Loading…
7 of 13 tasks
[PyTorch] [torch.compile] Prepare the custom-op framework and LayerNorm modules for compile wiring
#44 opened Sep 3, 2026 by pggPL Owner Loading…
6 of 8 tasks
[PyTorch] [torch.compile] torch.compile support for LayerNormLinear and LayerNormMLP
#43 opened Sep 3, 2026 by pggPL Owner Loading…
7 of 8 tasks
[PyTorch] [torch.compile] Prepare LayerNormLinear and LayerNormMLP for torch.compile
#42 opened Sep 3, 2026 by pggPL Owner Loading…
6 of 8 tasks
[PyTorch] Fix fake kernel metadata mismatch for empty-input MoE permute ops
#40 opened Aug 31, 2026 by pggPL Owner Loading…
7 of 13 tasks
[PyTorch] Schedule delayed-scaling updates after backward
#39 opened Aug 20, 2026 by pggPL Owner Loading…
9 of 13 tasks
[PyTorch] Delegate delayed-scaling updates to RecipeState
#38 opened Aug 20, 2026 by pggPL Owner Loading…
8 of 13 tasks
Loosen fp16 atol in ONNX export tests for onnxruntime 1.29 CPU numerics
#37 opened Aug 20, 2026 by pggPL Owner Loading…
7 of 13 tasks
[PyTorch] DeepSeekV3Layer: full MoE transformer layer (MLA + DeepSeek MoE)
#36 opened Aug 18, 2026 by pggPL Owner • Draft
7 of 13 tasks
CI: minimum supported PyTorch check
#35 opened Aug 18, 2026 by pggPL Owner Loading…
[PyTorch] Finalize quantization updates at logical backward boundaries
#34 opened Aug 17, 2026 by pggPL Owner • Draft
7 of 8 tasks
[PyTorch] [torch.compile] torch.compile support for GroupedLinear
#32 opened Aug 13, 2026 by pggPL Owner • Draft
7 of 13 tasks
[WIP] Torch stable ABI foundation: torch_compat layer + stable Tensor pybind caster
#31 opened Aug 10, 2026 by pggPL Owner • Draft
7 of 13 tasks
[PyTorch] Declare Bias and the activations as custom ops
#30 opened Aug 5, 2026 by pggPL Owner • 2/2 Loading…
7 of 13 tasks
[PyTorch] torch.compile for an OperationFuser group holding one operation
#29 opened Aug 5, 2026 by pggPL Owner • 1/2 Loading…
7 of 13 tasks
Carry symbolic shapes through compile custom ops community-contribution
#26 opened Aug 4, 2026 by shino16 Loading…
13 tasks
[PyTorch][CI] Fix test issues: connect test files that are not wired into qa
#25 opened Jul 30, 2026 by pggPL Owner Loading…
8 of 13 tasks
[JAX] Fix typo in qkv_layout docstring: seperate -> separate ai-approved Reviewed by ai_pr_bot and judged trivially simple
#23 opened Jul 27, 2026 by pggPL Owner Loading…
[PyTorch][torch.compile] Support for DotProductAttention on flash and unfused backends
#22 opened Jul 27, 2026 by pggPL Owner Loading…
9 tasks done
[PyTorch] Centralize TE↔torch binary boundary in a single facade file
#21 opened Jul 17, 2026 by pggPL Owner Loading…
5 of 6 tasks
[PyTorch] torch.compile support for UnfusedDotProductAttention
#19 opened Jul 8, 2026 by pggPL Owner Loading…
7 of 13 tasks
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.