Lightning-AI/pytorch-lightning

View on GitHub

Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

Stars: 31.4k
Forks: 3.8k
Open beginner issues: 5
Indexed issues: 849
Avg merge: 6d 7h
Merged PRs (30d): 6

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 17, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: help wanted good first issue

Beginner friendly Everything

Open All

849 open issues indexed

First epoch after resuming from checkpoint uses sampler epoch 0 (regression from #20775, num_workers > 0) Open

bug needs triage ver: 2.6.x

Difficulty 3/5 1-2 days Newbie friendliness 76/100

Lightning-AI/pytorch-lightning#21938 · Opened Sep 8, 2026
test_tensorboard_with_symlink fails on Windows without symlink privilege Open

bug logger: tensorboard tests ver: 2.7.x

Difficulty 2/5 1-3 hours Newbie friendliness 88/100

Lightning-AI/pytorch-lightning#21932 · 1 comment · Opened Sep 6, 2026
[Fabric] FSDPStrategy ignores error_if_nonfinite in gradient clipping Open

Difficulty 3/5 1-2 days Newbie friendliness 70/100

Lightning-AI/pytorch-lightning#21928 · 1 comment · Opened Sep 3, 2026
[mobile iOS Safari] Hyperlinks from the main docs page all give "Version not found" Open

docs needs triage

Difficulty 3/5 1-2 days Newbie friendliness 55/100

Lightning-AI/pytorch-lightning#21911 · 2 comments · Opened Aug 19, 2026
Unified storage_options support for FSDPStrategy, ModelParallelStrategy, and TorchCheckpointIO Open

feature needs triage

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21905 · Opened Aug 17, 2026
scale_batch_size runs zero batches per trial when started after training has begun Open

ver: 2.7.x

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21903 · 1 comment · Opened Aug 14, 2026
BaseFinetuning re-freezes already unfrozen modules when resuming from a checkpoint Open

ver: 2.7.x

Difficulty 3/5 1-2 days Newbie friendliness 25/100

Lightning-AI/pytorch-lightning#21901 · 1 comment · Opened Aug 14, 2026
Many time passes between epochs Open

bug needs triage ver: 2.2.x ver: 2.4.x

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21895 · 3 comments · Opened Aug 11, 2026
`CombinedLoader(max_size_cycle)` stops immediately when one iterable is empty Open

ver: 2.7.x

Difficulty 3/5 1-2 days Newbie friendliness 52/100

Lightning-AI/pytorch-lightning#21893 · Opened Aug 9, 2026
Optimize remote checkpoint loading with parallel multiprocess downloads and zero-copy mmap Open

feature needs triage

Difficulty 5/5 Over a week Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21868 · Opened Jul 30, 2026
compatibility with rich 15.0.0 Open

bug progress bar: rich ver: 2.6.x

Difficulty 3/5 1-2 days Newbie friendliness 58/100

Lightning-AI/pytorch-lightning#21865 · Opened Jul 29, 2026
sync_dist mean-reduced epoch metrics are inflated when ranks log unequal batch counts (integer batch-size accumulator floors during mean-sync) Open

bug logger ver: 2.6.x ver: 2.7.x

Difficulty 2/5 1-3 hours Newbie friendliness 25/100

Lightning-AI/pytorch-lightning#21859 · 2 comments · Opened Jul 23, 2026
ModelCheckpoint: manual-optimization pre-step snapshot is not applied when the save is deferred to on_validation_end Open

callback: model checkpoint ver: 2.6.x ver: 2.7.x

Difficulty 3/5 1-2 days Newbie friendliness 74/100

Lightning-AI/pytorch-lightning#21842 · 3 comments · Opened Jul 17, 2026
Mixed precision autocast is disabled when running FSDP on CPU Open

bug ver: 2.5.x ver: 2.6.x

Difficulty 3/5 1-2 days Newbie friendliness 68/100

Lightning-AI/pytorch-lightning#21815 · Opened Jul 9, 2026
Device mesh initialization fails when running FSDP on CPU Open

bug ver: 2.5.x ver: 2.6.x

Difficulty 2/5 1-3 hours Newbie friendliness 74/100

Lightning-AI/pytorch-lightning#21814 · Opened Jul 9, 2026
`ModelCheckpoint` deletes *previous run's* checkpoint when remote filesystem Open

bug callback: model checkpoint ver: 2.6.x

Difficulty 3/5 1-2 days Newbie friendliness 70/100

Lightning-AI/pytorch-lightning#21813 · Opened Jul 8, 2026
Enable FSDP strategy on CPU (accelerator="cpu") Open

feature strategy: fsdp

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21812 · Opened Jul 8, 2026
Add an end-to-end anomaly detection tutorial using PyTorch Lightning Open

docs

Difficulty 5/5 Over a week Newbie friendliness 45/100

Lightning-AI/pytorch-lightning#21809 · 7 comments · Opened Jul 6, 2026
Graceful stop on an extrenal file marker Open

callback feature

Difficulty 3/5 1-2 days Newbie friendliness 52/100

Lightning-AI/pytorch-lightning#21795 · 1 comment · Opened Jul 1, 2026
WandbLogger marks crashed runs as "Finished" instead of "Failed" in W&B sweeps Open

bug logger: wandb ver: 2.7.x

Difficulty 3/5 1-2 days Newbie friendliness 65/100

Lightning-AI/pytorch-lightning#21787 · Opened Jun 28, 2026
Log key prefix controls for Trainer, WandbLogger, and CometLogger Open

feature

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21782 · Opened Jun 28, 2026
Allow epochs with different size Open

feature waiting on author

Difficulty 4/5 3-5 days Newbie friendliness 45/100

Lightning-AI/pytorch-lightning#21773 · 4 comments · Opened Jun 23, 2026
Transformer Engine plugin fails to check weight exists for LayerNorm Open

bug ver: 2.6.x

Difficulty 2/5 1-3 hours Newbie friendliness 72/100

Lightning-AI/pytorch-lightning#21755 · 8 comments · Opened Jun 3, 2026
EMA validation swap models before first update_parameters Open

bug ver: 2.6.x

Difficulty 2/5 1-3 hours Newbie friendliness 68/100

Lightning-AI/pytorch-lightning#21724 · Opened May 19, 2026
AsyncCheckpointIO Should Clone() to CPU not on GPU Open

bug ver: 2.7.x

Difficulty 2/5 1-3 hours Newbie friendliness 55/100

Lightning-AI/pytorch-lightning#21630 · 1 comment · Opened Mar 30, 2026
Help PyG to support convertig homogenous LightningModule GNNs into heterogenous ones Open

feature

Difficulty 5/5 Over a week Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21629 · 1 comment · Opened Mar 30, 2026
CometLogger does not write logs in offline mode Open

bug ver: 2.5.x

Difficulty 1/5 Under an hour Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21617 · Opened Mar 27, 2026
Feature Request : Gradient Statistics Monitoring Callback Open

callback feature

Difficulty 5/5 Over a week Newbie friendliness 25/100

Lightning-AI/pytorch-lightning#21589 · Opened Mar 17, 2026
The example_input_array in LightningModule does not support using NestedTensor with layout=torch.jagged within the model. Open

bug needs triage ver: 2.5.x

Difficulty 3/5 1-2 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21588 · Opened Mar 17, 2026
Group the metrics in Tensorboard Open

bug logger: tensorboard ver: 2.5.x

Difficulty 5/5 Over a week Newbie friendliness 30/100

Lightning-AI/pytorch-lightning#21578 · 2 comments · Opened Mar 10, 2026
Frozen dataclass as a batch Open

feature

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21577 · 6 comments · Opened Mar 10, 2026
Cross-validation while still enabling CLI control Open

feature lightningcli

Difficulty 5/5 Over a week Newbie friendliness 30/100

Lightning-AI/pytorch-lightning#21574 · Opened Mar 9, 2026
Revamp Async Checkpointing to Use PyTorch Distributed Checkpoint (DCP) async_save Open

callback: model checkpoint checkpointing feature

Lightning-AI/pytorch-lightning#21557 · 1 assignee · Opened Feb 27, 2026
Set experiment tags in MLflow Open

feature logger: mlflow

Difficulty 3/5 1-2 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21554 · Opened Feb 26, 2026
Epoch start freezing regression Open

bug ver: 2.5.x ver: 2.6.x

Difficulty 4/5 3-5 days Newbie friendliness 30/100

Lightning-AI/pytorch-lightning#21550 · 3 comments · 1 reaction · Opened Feb 25, 2026
Cannot make test case unused parameters for proposed strategy in PR Open

bug ver: 2.5.x

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21548 · 2 comments · Opened Feb 24, 2026
Refactor throughput dtype inference into the Precision plugin API Open

refactor

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21535 · 1 comment · 1 reaction · Opened Feb 11, 2026
Add stepwise scheduling for GradientAccumulationScheduler Open

callback: gradient accumulation feature

Difficulty 4/5 3-5 days Newbie friendliness 45/100

Lightning-AI/pytorch-lightning#21534 · 2 comments · Opened Feb 11, 2026
`BackboneFinetuning`: `train_bn` only applied during unfreezing phase Open

needs triage refactor

Difficulty 4/5 3-5 days Newbie friendliness 45/100

Lightning-AI/pytorch-lightning#21531 · 1 comment · 1 reaction · Opened Feb 9, 2026
`Fabric.all_reduce` silently fails when tensors are on CPU Open

docs fabric

Lightning-AI/pytorch-lightning#21530 · 4 comments · 1 assignee · Opened Feb 7, 2026
Tensorboard logging breaks with certain scalar values with numpy >= 2.4.0 Open

bug logger logger: tensorboard ver: 2.5.x

Difficulty 2/5 1-3 hours Newbie friendliness 42/100

Lightning-AI/pytorch-lightning#21503 · 1 comment · 1 reaction · Opened Jan 23, 2026
Dataloader reload bug when loading from checkpoint Open

bug checkpointing lightningdatamodule ver: 2.5.x

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21492 · 4 comments · Opened Jan 16, 2026
forced CPU-GPU optimization at the end of every training step resulting in underutilization of GPU Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Lightning-AI/pytorch-lightning#21487 · 2 comments · 1 reaction · Opened Jan 14, 2026
LightningModule -Inference in Research- phrasing issue Open

docs waiting on author

Difficulty 1/5 Under an hour Newbie friendliness 55/100

Lightning-AI/pytorch-lightning#21472 · 1 comment · Opened Jan 5, 2026
Trackio logger Open

feature logger

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21466 · 2 comments · 2 reactions · Opened Jan 2, 2026
[Doc] https://lightning.ai/docs/pytorch/stable/ not showing the latest 2.6.0 doc Open

bug docs

Lightning-AI/pytorch-lightning#21465 · 4 comments · 1 assignee · Opened Jan 1, 2026
Manual `Dataloader` reloading during training Open

feature needs triage

Difficulty 5/5 Over a week Newbie friendliness 30/100

Lightning-AI/pytorch-lightning#21448 · 3 reactions · Opened Dec 23, 2025
Lightning sets shuffle=True for custom samplers when wrapping them for distributed sampling Open

bug distributed ver: 2.5.x

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21447 · Opened Dec 22, 2025
Fix broken examples exploration link in README (404 Error) Open

docs

Difficulty 1/5 Under an hour Newbie friendliness 52/100

Lightning-AI/pytorch-lightning#21445 · Opened Dec 22, 2025
MLflow model format required for the Register Model Open

feature logger: mlflow

Difficulty 5/5 Over a week Newbie friendliness 35/100

Lightning-AI/pytorch-lightning#21437 · Opened Dec 20, 2025

Lightning-AI/pytorch-lightning

Get new issues in your inbox