Qwen3-Omni Checkpoint Mapping

[entrpn]

Original author @eitanporat in #2728

Description

Checkpoint conversion mapping from hf

Tests

• Converted HF weights to orbax checkpoint.
• Tested regressions on other multimodal models by running SFT using tests/end_to_end/tpu/gemma3/4b/test_gemma3_multimodal_sft.sh (did not run jetstream/generation).
• Logs: https://paste.googleplex.com/5213472762757120

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.

[codecov]

Codecov Report

❌ Patch coverage is 0% with 195 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...xText/utils/ckpt_conversion/utils/param_mapping.py	0.00%	185 Missing ⚠️
src/MaxText/layers/encoders.py	0.00%	7 Missing ⚠️
src/MaxText/layers/models.py	0.00%	2 Missing ⚠️
src/MaxText/utils/ckpt_conversion/utils/utils.py	0.00%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

[hengtaoguo]

I wonder if you have tried the full checkpoint conversion? Are all layers filled with weights? Not sure if its ready to decode now, seems we are pending two more PRs.