Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Pull requests: pytorch/ao

New pull request New
316 Open 3,246 Closed
316 Open 3,246 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

[mxfp8 moe training] add support for bias parm in _to_mxfp8_then_scaled_grouped_mm CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: training quantize_ api training flow moe mx
#4386 opened May 12, 2026 by danielvegamyhre Contributor Loading…
1
1
[test][xpu] Port more test/quantization/quantize_/workflows/ cases to Intel GPU CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4384 opened May 12, 2026 by zxd1997066 Contributor • Draft
1
Add pading in linear dispatch of Float8Sparse2x4_1DData1DMetadataTensor (#4382) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported meta-exported module: rocm
#4382 opened May 11, 2026 by bbeckca Contributor Loading…
3
[pat] (5/5) Add NMSparseConstraint and MinSparsityConstraint CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: not user facing Use this tag if you don't want this PR to show up in release notes
#4381 opened May 11, 2026 by lisjin Contributor Loading…
3
[pat] (4/5) Simplify iterative reweight test cases CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: not user facing Use this tag if you don't want this PR to show up in release notes
#4380 opened May 11, 2026 by lisjin Contributor Loading…
3
[pat] (3/5) Add KElementGrouper CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: not user facing Use this tag if you don't want this PR to show up in release notes
#4379 opened May 11, 2026 by lisjin Contributor Loading…
5
[pat] (2/5) Add QKGrouper and refactor AttentionHeadGrouper for vmap CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: not user facing Use this tag if you don't want this PR to show up in release notes
#4378 opened May 11, 2026 by lisjin Contributor Loading…
3
[pat] (1/5) DTensor proxmap fix + low-rank speedups CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: not user facing Use this tag if you don't want this PR to show up in release notes
#4377 opened May 11, 2026 by lisjin Contributor Loading…
3
[test][xpu] Fix pip issue in Intel GPU CI ciflow/xpu label used to trigger xpu CI jobs CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4375 opened May 10, 2026 by zxd1997066 Contributor Loading…
1
[nvfp4_training] Support NVFP4 linear training on Blackwell CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4369 opened May 7, 2026 by rdspring1 Contributor Loading…
3
test low precision attention api for fp8 cudnn per tensor CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: inference quantize_ api inference flow
#4367 opened May 5, 2026 by howardzhang-cv Contributor • Draft
1
add inplace method for low precision attention CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: inference quantize_ api inference flow
#4366 opened May 5, 2026 by howardzhang-cv Contributor Loading…
1
Fix regex escape characters in documentation CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4363 opened May 1, 2026 by DrStone1971 Loading…
6
Fix INT8 quantization precision loss for bf16 inputs in IntxUnpackedToInt8Tensor CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: core changes affecting multiple modules, e.g. base config/tensor, observers, quant ops
#4362 opened May 1, 2026 by mergennachin Contributor • Draft
3
fix(setup.py): exclude top-level test/ from find_packages CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4361 opened May 1, 2026 by Anai-Guo Contributor Loading…
1
3
[CI Fix] Update float8 saturation test for PyTorch nightly behavior CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4359 opened May 1, 2026 by jainapurva Contributor Loading…
1
Flydsl mxfp8 quantize CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4357 opened Apr 30, 2026 by zstreet87 Loading…
6
[xpu][test] Enable float8 dtensor tests for Intel XPU ciflow/xpu label used to trigger xpu CI jobs CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: not user facing Use this tag if you don't want this PR to show up in release notes
#4354 opened Apr 29, 2026 by arlesniak Contributor • Draft
2
[xpu][test] Enable MoE distributed unit tests for Intel XPU CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4353 opened Apr 29, 2026 by kgajdamo • Draft
2
feat(sparsity): Add FisherPruner — FIM-guided weight pruning with calibration-based eFIM accumulation CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4352 opened Apr 28, 2026 by ramkrishs • Draft
8
gptq: weight samples by tokens instead of batches CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4351 opened Apr 28, 2026 by vkuzo Contributor Loading…
2
add more options to gptq moe script CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4350 opened Apr 28, 2026 by vkuzo Contributor Loading…
2
[xpu][test] Enable MoE fp8/mxfp8/nvfp4 tests for Intel XPU CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4349 opened Apr 28, 2026 by kgajdamo • Draft
2
Fix symmetric quantization scale to match C++ ChooseQuantizationParams (#180990) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported meta-exported
#4345 opened Apr 28, 2026 by fantaosha Loading…
3
[xpu][test] Enable MoE MX format kernels and tensor tests for Intel XPU ciflow/xpu label used to trigger xpu CI jobs CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: core changes affecting multiple modules, e.g. base config/tensor, observers, quant ops
#4342 opened Apr 27, 2026 by kgajdamo • Draft
3
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.