Search results

ConstantLR accepts factor=0 and crashes with ZeroDivisionError after total_iters steps

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

module: error checkingBugs related to incorrect/lacking error checkingBugs related to incorrect/lacking error checking

module: LrScheduler

module: optimizerRelated to torch.optimRelated to torch.optim

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#185006 In pytorch/pytorch;

· Sriniketh24 opened on May 23, 2026

torch.export.export accepts a parameter whose storage is too small, then save/runtime fails with low-level setStorage error

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

oncall: export

oncall: pt2

Status: Open.

#184996 In pytorch/pytorch;

· FagonViendra opened on May 23, 2026

torch.compile crashes when torch.cat uses axis= keyword — sink_cat_after_pointwise pass

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

module: inductor

module: numpyRelated to numpy support, and also numpy compatibility of our operatorsRelated to numpy support, and also numpy compatibility of our operators

oncall: pt2

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184992 In pytorch/pytorch;

· wuyii8941 opened on May 23, 2026

deprecation of derived quantized dtypes (quint8, qint8, qint32)

Status: Open.

#184982 In pytorch/pytorch;

· vkuzo opened on May 23, 2026

[torch.compile] AOTAutograd fails to coerce plain Tensor tangent for expected AsyncCollectiveTensor

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

has workaround

module: aotdispatchumbrella label for AOTAutograd issuesumbrella label for AOTAutograd issues

module: dtensordistributed tensor tagdistributed tensor tag

module: pt2-dispatcherPT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op,PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op,

oncall: distributedAdd this issue/PR to distributed oncall triage queueAdd this issue/PR to distributed oncall triage queue

oncall: pt2

tensor subclassRelated to tensor subclassesRelated to tensor subclasses

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184980 In pytorch/pytorch;

· PAULBRYANT8 opened on May 23, 2026

[Inductor] Codegen Gluon

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

featureA request for a proper, new feature.A request for a proper, new feature.

module: inductor

oncall: pt2

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184903 In pytorch/pytorch;

· liqiangxl opened on May 22, 2026

ROCm: Python 3.15 wheel builds blocked by composable_kernel using removed importlib.abc.Loader.load_module()

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

module: buildBuild system issuesBuild system issues

module: ciRelated to continuous integrationRelated to continuous integration

module: dependency bugProblem is not caused by us, but caused by an upstream library we useProblem is not caused by us, but caused by an upstream library we use

module: python versionIssues related to specific Python versionsIssues related to specific Python versions

module: rocmAMD GPU support for PytorchAMD GPU support for Pytorch

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184900 In pytorch/pytorch;

· atalman opened on May 22, 2026

XPU: Triton wheel builds for Python 3.15 are currently broken; skipped in build-triton-wheel.yml

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

module: ciRelated to continuous integrationRelated to continuous integration

module: python versionIssues related to specific Python versionsIssues related to specific Python versions

module: xpuIntel XPU related issuesIntel XPU related issues

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184901 In pytorch/pytorch;

· atalman opened on May 22, 2026

[Test] functorch/test_vmap.py RuntimeError during test setup with CPU and CUDA

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

module: functorchPertaining to torch.func or pytorch/functorchPertaining to torch.func or pytorch/functorch

module: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn't

module: testsIssues related to tests (not the torch.testing module)Issues related to tests (not the torch.testing module)

module: vmap

triage review

Status: Open.

#184894 In pytorch/pytorch;

· gderossi opened on May 22, 2026

[Inductor] argmin/min returns incorrect indices for boolean tensors

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

module: correctness (silent)issue that returns an incorrect result silentlyissue that returns an incorrect result silently

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

module: inductor

module: sorting and selection

oncall: pt2

triage review

Status: Open.

#184893 In pytorch/pytorch;

· Flamefire opened on May 22, 2026

vmap of autograd.grad fails on the output of model run on AMP

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

has workaround

module: amp (automated mixed precision)autocastautocast

module: autogradRelated to torch.autograd, and the autograd engine in generalRelated to torch.autograd, and the autograd engine in general

module: functorchPertaining to torch.func or pytorch/functorchPertaining to torch.func or pytorch/functorch

module: norms and normalization

module: vmap

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184890 In pytorch/pytorch;

· ValerianRey opened on May 22, 2026

Clarify torch.searchsorted behavior when sorted_sequence is unsorted

bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bot

enhancementNot as big of a feature, but technically not a bug. Should be easy to fixNot as big of a feature, but technically not a bug. Should be easy to fix

module: docsRelated to our documentation, both in docs/ and docblocksRelated to our documentation, both in docs/ and docblocks

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#184883 In pytorch/pytorch;

· FagonViendra opened on May 22, 2026