Issues
is:issue state:open
is:issue state:open
Search results
- Status: Open.#169752 In pytorch/pytorch;
ReplicationPad2dwith extremely large padding value causes segmentation fault
module: crashProblem manifests as a hard crash, as opposed to a RuntimeErrorProblem manifests as a hard crash, as opposed to a RuntimeErrormodule: nnRelated to torch.nnRelated to torch.nntriagedThis 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 moduleStatus: Open.#169741 In pytorch/pytorch;[MPS][Inductor] AdaptiveMaxPool{1,2}d produces incorrect results (numerical correctness)
module: correctness (silent)issue that returns an incorrect result silentlyissue that returns an incorrect result silentlymodule: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders frameworkStatus: Open.#169738 In pytorch/pytorch;CI runner runs out of disk space when building Sphinx site for site branch in pytorch/docs
module: doc infraRelated to pytorch.org/docs, deployment of, and servingRelated to pytorch.org/docs, deployment of, and servingmodule: docsRelated to our documentation, both in docs/ and docblocksRelated to our documentation, both in docs/ and docblocksmodule: infraRelates to CI infrastructureRelates to CI infrastructureStatus: Open.#169728 In pytorch/pytorch;[dynamo] Show where in user code side effects are happening
dynamo-triage-dec2025small/mid-sized dynamo tasks that we would like to be completed in the near futuresmall/mid-sized dynamo tasks that we would like to be completed in the near futuremodule: compile uxUX issues related to torch.compileUX issues related to torch.compiletriagedThis 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 moduleStatus: Open.#169722 In pytorch/pytorch;[CPU] Stride 3 bfloat16 convolutions appear incorrect
module: convolutionProblems related to convolutions (THNN, THCUNN, CuDNN)Problems related to convolutions (THNN, THCUNN, CuDNN)module: cudaRelated to torch.cuda, and CUDA support in generalRelated to torch.cuda, and CUDA support in generalneeds reproductionEnsure you have actionable steps to reproduce the issue. Someone else needs to confirm the repro.Ensure you have actionable steps to reproduce the issue. Someone else needs to confirm the repro.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 moduleStatus: Open.#169688 In pytorch/pytorch;ROCM nightly builds are broken since Dec 1 , 2025
module: rocmAMD GPU support for PytorchAMD GPU support for PytorchrocmThis tag is for PRs from ROCm teamThis tag is for PRs from ROCm teamrocm priorityhigh priority ROCm PRs from performance or other aspectshigh priority ROCm PRs from performance or other aspectsStatus: Open.#169684 In pytorch/pytorch;Header-hiding when TORCH_TARGET_VERSION is set is broken for editable torch installs
module: buildBuild system issuesBuild system issuesmodule: cpp-extensionsRelated to torch.utils.cpp_extensionRelated to torch.utils.cpp_extensiontriagedThis 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 moduleStatus: Open.#169675 In pytorch/pytorch;- Status: Open.#169672 In pytorch/pytorch;
[inductor] TIMM model training cudagraphs regression
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 moduleStatus: Open.#169671 In pytorch/pytorch;Energy normalization error on torch.fft.fft2 and torch.fft.ifft2 when the matrix size = [2048, 2048]
Status: Open.#169670 In pytorch/pytorch;Inconsistent<code>flop_counter.get_total_flops()</code> results for a custom decoder model
oncall: profilerprofiler-related issues (cpu, gpu, kineto)profiler-related issues (cpu, gpu, kineto)Status: Open.#169666 In pytorch/pytorch;