Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
201 Open 2,488 Closed
201 Open 2,488 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Include test_assets from GCS in tpu/gpu dockerfiles
#2898 opened Dec 29, 2025 by SurbhiJainUSC Loading…
4 tasks done
4
Fix SFT + GA (gradient accumulation)
#2897 opened Dec 29, 2025 by gobbleturk Loading…
4 tasks done
1
Small updates on MoE configuration gemini-review
#2895 opened Dec 29, 2025 by RissyRan Loading…
4 tasks done
@richjames0
6
Fix Pydantic validation error for Vertex Tensorboard args
#2893 opened Dec 29, 2025 by hsuan-lun-chiang Loading…
4 tasks done
3
Update loss free & loss control load balance gemini-review pull ready
#2892 opened Dec 27, 2025 by RissyRan Loading…
4 tasks done
14
Add fractional batch size search for MaxText estimator draft Draft PR
#2891 opened Dec 27, 2025 by NuojCheng • Draft
4 tasks
1
Feat: Update RL on Multi-Host TPUs tutorial for clarity and structure
#2890 opened Dec 24, 2025 by RexBearIU Loading…
4 tasks done
13
Fix Codecov setup
#2889 opened Dec 24, 2025 by shralex Loading…
4 tasks done
1
Posttraining documentation refine
#2888 opened Dec 24, 2025 by hsuan-lun-chiang Loading…
4 tasks done
11
fix: add missing module init and update docs
#2887 opened Dec 24, 2025 by mesakhcienet • Draft
4 tasks done
Implement Best Fit Packing Algorithm in Grain Pipeline for Reduced Padding
#2886 opened Dec 24, 2025 by bzantium Loading…
4 tasks done
1
1
feat: migrate pipeline to nnx
#2885 opened Dec 24, 2025 by mesakhcienet • Draft
4 tasks
Implement Warmup-Stable-Decay (WSD) Learning Rate Schedule
#2883 opened Dec 24, 2025 by bzantium Loading…
4 tasks done
1
1
SFT documentation refine - single TPU VM procedure
#2880 opened Dec 23, 2025 by xibinliu Loading…
4 tasks done
5
feat: support qwen3 next with muon
#2875 opened Dec 23, 2025 by Prayer3th Loading…
4 tasks done
@parambole @shuningjin @Rohan-Bierneni
6
feat: support KDA
#2874 opened Dec 23, 2025 by jimoosciuc Loading…
4 tasks done
@parambole @RissyRan @Rohan-Bierneni
[DECOUPLED-MODE] Adding Decoupling Logic
#2865 opened Dec 21, 2025 by gulsumgudukbay Loading…
4 tasks done
Add automatic API documentation generation
#2862 opened Dec 19, 2025 by melissawm Loading…
4 tasks done
Adding supported models to post_training_index.md pull ready
#2860 opened Dec 19, 2025 by jacoguzo Loading…
4 tasks done
1
Jacoguzo/tutorials editable code pull ready
#2856 opened Dec 18, 2025 by jacoguzo Loading…
4 tasks done
2
Fix two issues that blocks training loop with continuous checkpoint enabled.
#2854 opened Dec 17, 2025 by copybara-service bot Loading…
reduce tfds packing intermediate batch_size
#2847 opened Dec 16, 2025 by aireenmei • Draft
4 tasks
Docs: Enhance landing pages for Run, Guides, and Reference
#2839 opened Dec 16, 2025 by RobMulla Loading…
4 tasks done
1
Docs streamline fixes
#2838 opened Dec 16, 2025 by RobMulla Loading…
4 tasks done
Migrate Qwen3 Next to Checkpoint Util
#2832 opened Dec 15, 2025 by Rohan-Bierneni Loading…
4 tasks done
@Rohan-Bierneni
ProTip! no:milestone will show everything without a milestone.