Skip to content

Conversation

@Rohan-Bierneni
Copy link
Collaborator

@Rohan-Bierneni Rohan-Bierneni commented Dec 15, 2025

Description

Qwen3 Next currenty has checkpoint scripts for scanned and unscanned from HF to Maxtext. This pr onboards the model to the checkpoint util which allows for bidirectional conversion between HF and MaxText.

If the change fixes a bug or a Github issue, please include a link, e.g.,:
FIXES: b/123456

Tests

Will run forward pass logit checker on converted checkpoint from Maxtext -> HF -> Maxtext for scanned and post results here:

Current status:

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.

PiperOrigin-RevId: 845968954

Added initial qwen3 next migration to checkpoint util

revert changes to api server script

Update hooks function for bidirectional conversion

Add functions for qwen3 next to param mapping

Remove check for hf to maxtext conversion

HF -> Maxtext scanned conversion works now

revert port forward script to main
@Rohan-Bierneni Rohan-Bierneni force-pushed the rbierneni-checkpoint-util branch from 3a662c6 to 78b86ec Compare December 18, 2025 05:05
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.

2 participants