Search results

torch.export.export Fails with RuntimeError and "Please file an issue on github"

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

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.

#188153 In pytorch/pytorch;

· michaelheider opened on Jun 25, 2026

[vllm] [2.13 regression] block-scaled fp8 MoE numerical mismatch (Kernels MoE / FusedMoE) - assert_close fails after #187484 compile revert

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: floatx (formerly float8)For torch.float8_e5m2 and torch.float8_e4m3 and other sub 8-bit float typesFor torch.float8_e5m2 and torch.float8_e4m3 and other sub 8-bit float types

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

module: vllm

oncall: pt2

triage review

Status: Open.

#188152 In pytorch/pytorch;

· atalman opened on Jun 25, 2026

torch.cond fails in torch.compile with simpliest symbolic shapes (torch 2.12+)

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: dynamic shapes

module: dynamo

module: higher order operatorstorch.cond and similartorch.cond and similar

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: 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.

#188151 In pytorch/pytorch;

· cooldome opened on Jun 25, 2026

torch.compile reduce-overhead: CUDAGraphs recompiles on every batch with dynamic padding (HF training loop)

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

has workaround

module: compile uxUX issues related to torch.compileUX issues related to torch.compile

module: cuda graphsAbility to capture and then replay streams of CUDA kernelsAbility to capture and then replay streams of CUDA kernels

module: dynamic shapes

module: performanceIssues related to performance, either of kernel code or framework glueIssues related to performance, either of kernel code or framework glue

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.

#188150 In pytorch/pytorch;

· ajwharton opened on Jun 25, 2026

DISABLED test_custom_op_dce_dynamic_shapes_cuda (__main__.DynamicShapesCodegenGPUTests)

module: flaky-testsProblem is a flaky test in CIProblem is a flaky test in CI

module: inductor

oncall: pt2

skippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.

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.

#188144 In pytorch/pytorch;

· pytorch-bot opened on Jun 25, 2026

DISABLED test_compare_cpu_linalg_polar_xpu_float32 (__main__.TestCommonXPU)

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: linear algebraIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmulIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmul

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

module: xpuIntel XPU related issuesIntel XPU related issues

skippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.

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.

#188135 In pytorch/pytorch;

· etaf opened on Jun 25, 2026

Pallas CPU mishandles synthetic-base mutated aliased inputs

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

oncall: cpu inductorCPU Inductor issues for Intel team to triageCPU Inductor issues for Intel team to triage

oncall: pt2

Status: Open.

#188133 In pytorch/pytorch;

· jansel opened on Jun 25, 2026

DISABLED test_fallback_kernel_with_symexpr_output_xpu (__main__.AOTInductorTestABICompatibleGpu)

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: xpuIntel XPU related issuesIntel XPU related issues

oncall: pt2

skippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.

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.

#188131 In pytorch/pytorch;

· guangyey opened on Jun 25, 2026

[MPS] RuntimeError: Failed to create function state object for igammac_dense_float2_float2when using torch.complex64

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: complexRelated to complex number support in PyTorchRelated to complex number support in PyTorch

module: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders framework

module: specialFunctions with no exact solutions, analogous to those in scipy.specialFunctions with no exact solutions, analogous to those in scipy.special

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.

#188129 In pytorch/pytorch;

· christianahmed905 opened on Jun 25, 2026

asarray does not work for boolean dtype when using buffer protocol

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: python array apiIssues related to the Python Array APIIssues related to the Python Array API

module: python frontendFor issues relating to PyTorch's Python frontendFor issues relating to PyTorch's Python frontend

module: tensor creation

triage review

Status: Open.

#188128 In pytorch/pytorch;

· lucyleeow opened on Jun 25, 2026

torch.compile causes persistent RSS growth during repeated inference in PyTorch 2.9 with ROCM

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: memory usagePyTorch is using more memory than it should, or it is leaking memoryPyTorch is using more memory than it should, or it is leaking memory

module: performanceIssues related to performance, either of kernel code or framework glueIssues related to performance, either of kernel code or framework glue

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

module: rocmAMD GPU support for PytorchAMD GPU support for Pytorch

oncall: pt2

triage review

Status: Open.

#188127 In pytorch/pytorch;

· iMountTai opened on Jun 25, 2026

DISABLED test_index_cuda_with_cpu_int64_propagate_real_tensors (__main__.PropagateRealTensorsFakeTensorTest)

module: flaky-testsProblem is a flaky test in CIProblem is a flaky test in CI

module: meta tensors

module: rocmAMD GPU support for PytorchAMD GPU support for Pytorch

skippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.

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.

#188123 In pytorch/pytorch;

· pytorch-bot opened on Jun 25, 2026