PyTorch Lightning

license

Apache-2.0(OSI)

source

public(Lightning-AI/pytorch-lightning)

core-gated

ungated(no enterprise or ee path in the repository root and no paid tier described in the README

Apache-2.0, an OSI licence, with the full source public and the shipped code in this repository. The LICENSE body was read in full: the stock Apache-2.0 text, its core byte-identical to the canonical template, with nothing after END OF TERMS AND CONDITIONS but the standard appendix. No enterprise or ee path in the repository root and no paid tier described in the readme; lightning ai sells a hosted platform, which is a separate product alongside rather than a gated core.

• https://raw.githubusercontent.com/Lightning-AI/pytorch-lightning/master/LICENSE recorded 2026-08-31

  LICENSE body, read in full: the stock Apache License 2.0 text, byte-identical to the canonical template through END OF TERMS AND CONDITIONS and followed only by the standard appendix. No appended condition.

• https://raw.githubusercontent.com/Lightning-AI/pytorch-lightning/master/README.md recorded 2026-08-31

  README, read for the distribution and gating question: it documents pip-installing the library and running it locally, and describes no enterprise, paid or licence-gated build of the library. Lightning AI's hosted platform is a separate offering rather than a tier of this code.

Banded on the authoritative PyPI route: 12,410,527 downloads in the last 30 days for the pytorch-lightning package, which places it at level 5. The same repository also publishes an umbrella `lightning` package at 5,665,410 downloads over the same window; it is not declared as a second artifact because it redistributes this code alongside Lightning Fabric, so declaring it would claim an artifact that a separate product may later need, and the band is level 5 on either reading. The repository carries 31,313 stars, recorded only as corroboration.

• https://pypistats.org/api/packages/pytorch-lightning/recent recorded 2026-08-31

  last_month 12410527, last_week 2323920, last_day 238140 for the pytorch-lightning package.

• https://api.github.com/repos/Lightning-AI/pytorch-lightning recorded 2026-08-31

  Repo metadata for Lightning-AI/pytorch-lightning: stargazers_count 31313, archived false, pushed_at 2026-08-31.

Scored 4 as a library built on a framework rather than a framework. It structures and distributes a PyTorch training loop, which is the same shape as accelerate at 4 in this category, and depends on PyTorch for the tensor engine and autograd that put pytorch at 5. Its breadth of accelerator and strategy coverage is what holds it at the top of that band.

• https://raw.githubusercontent.com/Lightning-AI/pytorch-lightning/master/README.md recorded 2026-08-31

  The Key features section documents switching between CPU, GPU (CUDA and MPS), TPU, multi-GPU and multi-node execution with no code changes via the Fabric accelerator/devices/num_nodes arguments, and distributed strategies DDP, FSDP and DeepSpeed with mixed precision available out of the box.