Lightning-AI / Lightning-AI/pytorch-lightning

Intel GPU support?

Open
#20,938 5 comments 13 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

accelerator feature
Dominant language
Python
Stars
31.4k
Forks
3.8k
Avg merge
6d 7h
Merged PRs (30d)
6

Description

Dear Lightning team, @lantiga, @Borda,

Can you please prioritize Intel GPU support on your development roadmap?

My students and I have been waiting _years_ for Lightning support on Intel GPUs. I find it disappointing that this hardware remains unsupported despite being fully integrated in PyTorch itself since October 2024. Due to these continual delays, I have had to start removing Lightning from teaching materials, as well as discourage students from using it, for the simple reason that Lightning prevents students from using the latest hardware on our academic cluster here in Cambridge. More generally, I think not supporting features that are supported natively by PyTorch is frustrating and makes me question the value of higher-level frameworks. I think such frameworks should _expand_ support, if anything, not reduce it.

There have been various attempts such as #17700 #16834 #20349. I don't understand why these were closed and/or did not receive more internal support.

Could your team please prioritize this?

We are happy to help test a beta if that would speed things up.

cc @lantiga @borda @justusschock

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the prior Intel GPU discussions in issues #17700, #16834, and #20349, then compare Lightning's hardware support with PyTorch's Intel GPU support. The request is complete when Lightning has an agreed implementation scope and supports Intel GPUs, but this issue does not identify files, tests, or a concrete change.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.