Issues

• CPU peak memory calculation overestimates memory usage by 100x

  pytorch/benchmark#2707 · Filter by author Cynosure159 opened on Sep 16, 2026

• [huggingface] NameError: name 'AutoConfig' is not defined in download_model()

  pytorch/benchmark#2703 · Filter by author vinay-murakala opened on Sep 1, 2026

• HuggingFace models fail on Python 3.13: NameError from exec/eval in download_model (PEP 667)

  pytorch/benchmark#2698 · Filter by author amd-nishsoni opened on Aug 3, 2026

• Question: what is the intended sync relationship between pytorch/benchmarks/dynamo and benchmark/userbenchmark/dynamo/dynamobench?

  pytorch/benchmark#2675 · Filter by author zhili03 opened on Mar 19, 2026

• decoder_input_ids/decoder_inputs_embeds missing in example inputs for hf_T5_base

  pytorch/benchmark#2670 · Filter by author josefmalmstrom opened on Mar 9, 2026

• (auto)Bump diffusers version pin

  pytorch/benchmark#2655 · Filter by author StrongerXi opened on Nov 19, 2025
  1

• ModuleNotFoundError: No module named 'torchbenchmark.util.framework.fb'

  pytorch/benchmark#2653 · Filter by author jerly-hjzhou opened on Nov 14, 2025

• [Bug] ModuleNotFoundError: No module named 'torch._C._nativert' with recent PyTorch nightly

  pytorch/benchmark#2631 · Filter by author dingandchen opened on Aug 16, 2025
  1

• batching-a-multicategorical , spec, isssues

  pytorch/benchmark#2628 · Filter by author rsarpongstreetor opened on Aug 6, 2025

• Support for multiple nodes?

  pytorch/benchmark#2613 · Filter by author vsoch opened on May 2, 2025

• AttributeError: 'ModelTask' object has no attribute 'replace_invoke'

  pytorch/benchmark#2606 · Filter by author csc010228 opened on Apr 2, 2025

• [Userbenchmark] Failed to execute userbenchmark/distributed because Accelerate object has no attribute 'use_fp16'

  pytorch/benchmark#2593 · Filter by author aztecher opened on Feb 16, 2025

• Missing test_bench configuration support

  pytorch/benchmark#2590 · Filter by author shink opened on Feb 11, 2025
  1

• dlrm_s_pytorch.py profile api arguments mismatch

  pytorch/benchmark#2589 · Filter by author xpfjmj opened on Feb 8, 2025

• [release-test] GPU memory numbers are always 0 on A100 runner

  pytorch/benchmark#2574 · Filter by author xuzhao9 opened on Jan 22, 2025
  3

• [release-test] A100 3803mnist_hogwild-latency increase 10x on linux.aws.a100 vs [a100-runner]

  pytorch/benchmark#2573 · Filter by author atalman opened on Jan 21, 2025
  2

• [eager_fail_to_run] Fail to run torchrec_dlrm with --bfloat16

  pytorch/benchmark#2572 · Filter by author IvanKobzarev opened on Jan 17, 2025

• [eager_fail_to_run] cuda train sam

  pytorch/benchmark#2570 · Filter by author IvanKobzarev opened on Jan 17, 2025

• [eager_fail_to_run] resnet50_quantized_qat

  pytorch/benchmark#2569 · Filter by author IvanKobzarev opened on Jan 16, 2025

• [eager_fail_to_run] opacus_cifar10

  pytorch/benchmark#2568 · Filter by author IvanKobzarev opened on Jan 16, 2025

• [eager_fail_to_run ] mobilenet_v2_quantized_qat, eager run failed on dtype mismatch

  pytorch/benchmark#2567 · Filter by author IvanKobzarev opened on Jan 16, 2025

• Add SAM2 to TorchBench model set

  pytorch/benchmark#2566 · Filter by author huydhn opened on Jan 16, 2025
  4

  

• Inefficient usage of torch.randn() in vision/model_factory.py

  pytorch/benchmark#2555 · Filter by author perfNerdJK opened on Dec 11, 2024

• Benchmark reliability of torchbenchmarks

  pytorch/benchmark#2527 · Filter by author jerryzh168 opened on Oct 28, 2024
  4

• install_diffusers issue with stable_diffusion_xl

  pytorch/benchmark#2522 · Filter by author JasonMts opened on Oct 21, 2024