Skip to content

Add zero1 aot support in train compile#3872

Draft
NuojCheng wants to merge 1 commit into
mainfrom
chengnuojin-zero1-aot
Draft

Add zero1 aot support in train compile#3872
NuojCheng wants to merge 1 commit into
mainfrom
chengnuojin-zero1-aot

Conversation

@NuojCheng
Copy link
Copy Markdown
Collaborator

@NuojCheng NuojCheng commented May 11, 2026

Description

Previously AOT using ZeRO-1 sharding was not supported for train_compile. This PR debugs this issue and enable this funciton. It also adds a test protecting Zero-1 sharding for AOT.

Tests

CI and added test.

Checklist

Before submitting this PR, please make sure (put X in square brackets):

  • I have performed a self-review of my code. For an optional AI review, add the gemini-review label.
  • I have necessary comments in my code, particularly in hard-to-understand areas.
  • I have run end-to-end tests tests and provided workload links above if applicable.
  • I have made or will make corresponding changes to the doc if needed, including adding new documentation pages to the relevant Table of Contents (toctree directive) as explained in our documentation.

@NuojCheng NuojCheng force-pushed the chengnuojin-zero1-aot branch from 7f0683d to e884cb3 Compare May 11, 2026 21:28
@codecov
Copy link
Copy Markdown

codecov Bot commented May 11, 2026

Codecov Report

❌ Patch coverage is 40.00000% with 6 lines in your changes missing coverage. Please review.

Files with missing lines Patch % Lines
src/maxtext/trainers/pre_train/train_compile.py 40.00% 6 Missing ⚠️

📢 Thoughts on this report? Let us know!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant