Skip to content

Pull requests: pytorch/ao

New pull request New
470 Open 3,470 Closed
470 Open 3,470 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

Refactor MXTensor: align the swizzle execution with the configured swizzle setting 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: inference quantize_ api inference flow xpu Intel XPU related features
#4844 opened Aug 28, 2026 by xiaowangintel Collaborator Loading…
18
Refactor MXTensor: replace is_swizzled_scales bool with SwizzleType class hierarchy 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: inference quantize_ api inference flow xpu Intel XPU related features
#4843 opened Aug 28, 2026 by xiaowangintel Collaborator • Draft
1
docs: remove stale MXFP4 training entry CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4842 opened Aug 28, 2026 by GotFusion Contributor Loading…
1
1
Remove Float8 allow-in-graph custom autograd wrappers CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. meta-exported
#4840 opened Aug 27, 2026 by Jiapli930 Loading…
4
Make MXTensor subclass-friendly
#4839 opened Aug 27, 2026 by ahmadsarvmeily Loading…
8
[WIP][MX] Add FSDP2 support for inference weights CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4838 opened Aug 27, 2026 by GotFusion Contributor Loading…
1
1
[CI] Run only accelerator PT2E tests on GPU CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: pt2e_quant pt2 export quantization (prepare_pt2e, convert_pt2e, quantizer)
#4836 opened Aug 27, 2026 by iamzainhuda Contributor Loading…
6
[ROCm][CI][DO NOT MERGE] Test Run PT2E GPU for PR#4755 ciflow/rocm CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: rocm
#4835 opened Aug 26, 2026 by akashveramd Collaborator Loading…
@akashveramd
1
Fold away num_batches_tracked when the +1 is a lifted constant (#4834) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. meta-exported module: qat quantize_ api qat flow
#4834 opened Aug 26, 2026 by billmguo Contributor Loading…
15
Stop PruneOptimizer.__getattribute__ recursing (#4833) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. meta-exported module: optimizer quantized optimizers (torchao/optim)
#4833 opened Aug 26, 2026 by sibuachu Loading…
7
Limit MXFP8 PyTorch version guard to DTensor usage CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4831 opened Aug 26, 2026 by GotFusion Contributor Loading…
1
1
Refactor ToyConvModel into shared testing architectures CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4829 opened Aug 26, 2026 by AndyYin06 Loading…
1
Add opt-in KleidiAI support to BUCK (#4828) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. meta-exported module: inference quantize_ api inference flow
#4828 opened Aug 25, 2026 by JakeStevens Contributor Loading…
5
Fix generate_permute_indices max_len bounds CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4827 opened Aug 25, 2026 by gaoxiaomo Loading…
1
2
[pt2e][qat] Fix qparams mismatch in FusedMovingAvgObsFakeQuantize CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: pt2e_quant pt2 export quantization (prepare_pt2e, convert_pt2e, quantizer)
#4824 opened Aug 24, 2026 by andrewor14 Contributor Loading…
8
[CPU] Register the x86 inductor int8 lowering passes from quant_api CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4823 opened Aug 24, 2026 by Kanya-Mo Loading…
1
Gate MX/NVFP4 safetensors test configs on sm100 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
#4821 opened Aug 24, 2026 by alanhuangyoo Loading…
2
[mxfp8 training] Add cudnn wrapper for fused grouped GEMM + SwiGLU + MXFP8 quantization ops for MoE routed experts CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4820 opened Aug 24, 2026 by wolfcomos Loading…
@wolfcomos
3
Fix IndexError when excluding multiple FqnToConfig params with None CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4819 opened Aug 24, 2026 by gaurav-redhat Loading…
1 task done
1
1
Fix bitpacking default unpack dimension CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4816 opened Aug 23, 2026 by aswanth-07 Loading…
1
1
Refactor MXTensor: Use  SwizzleType  enum for device-agnostic scale layout dispatch 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: inference quantize_ api inference flow xpu Intel XPU related features
#4813 opened Aug 21, 2026 by xiaowangintel Collaborator Loading…
49
Fix _choose_qparams_affine collapsing to a scalar when block_size is all 1s CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4812 opened Aug 20, 2026 by amanyagami Loading…
1
1
Add scale/zero_point dimension validation to quant_primitives CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4811 opened Aug 20, 2026 by amanyagami Loading…
1
1
Fix wrong variable in test_handle_save_load_meta_impl error message CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4810 opened Aug 20, 2026 by amanyagami Loading…
1
1
Bounds-check block_size against packed tensors in Int4TilePackedTo4dTensor CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4809 opened Aug 20, 2026 by amanyagami Loading…
1
1
ProTip! Filter pull requests by the default branch with base:main.