Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
160 Open 3,552 Closed
160 Open 3,552 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

Add Qwen3.5 vision encoder and connector
#3962 opened May 21, 2026 by hengtaoguo Collaborator Loading…
4 tasks done
1
docs: clarify multi-host prerequisites (IAM roles and GKE auth plugin)
#3961 opened May 20, 2026 by darisoy Collaborator Loading…
4 tasks done
2
Improve RL parallelism ValueError message and UXR documentation
#3960 opened May 20, 2026 by darisoy Collaborator Loading…
4 tasks done
7
Update Jupyter Notebooks to convert MaxText checkpoint to Hugging Face
#3959 opened May 20, 2026 by YixuanWang-99 Collaborator Loading…
4 tasks done
14
Migrate remaining tests off tf
#3957 opened May 20, 2026 by aireenmei Collaborator • Draft
4 tasks done
1
Add Multi-tiered Checkpointing Support to Pathways
#3952 opened May 20, 2026 by SujeethJinesh Collaborator Loading…
4 tasks done
1
Remove tf from post-training
#3951 opened May 20, 2026 by aireenmei Collaborator Loading…
4 tasks done
1
Apply chat templating for lm_eval mmlu post training checkpoints
#3950 opened May 20, 2026 by dipannita08 Collaborator Loading…
4 tasks done
1
Add option to use Tokamax attention API in MaxText.
#3949 opened May 19, 2026 by copybara-service Bot Loading…
2
Fix MaxText Paged Attention OOM and AOT state freezing
#3948 opened May 19, 2026 by darisoy Collaborator • Draft
4 tasks done
1
Fix TPU kernel OOB memory access and correctness issues
#3946 opened May 19, 2026 by darisoy Collaborator Loading…
4 tasks done
7
NNX migration
#3943 opened May 19, 2026 by hsuan-lun-chiang Collaborator • Draft
4 tasks
1
Qwen3-30b learn to init
#3942 opened May 19, 2026 by gagika Collaborator Loading…
4 tasks done
4
Deprecate cloud_tpu_diagnostics support and update requirements lockfile
#3940 opened May 19, 2026 by SurbhiJainUSC Collaborator Loading…
4 tasks done
1
Add num_trays_per_compute_node.
#3937 opened May 18, 2026 by copybara-service Bot Loading…
1
Rename save/load_pytree to save/load. Eliminate most user-facing "pytree" terminology in favor of "state" as a more specific term. Add deprecations.py for handling deprecated public functions.
#3934 opened May 18, 2026 by copybara-service Bot Loading…
1
Supports batching in vLLM offline inference script pull ready
#3932 opened May 18, 2026 by learning-to-play Collaborator Loading…
4 tasks done
@jrplatin
1
Refactor tensor handling and add dynamic SafeTensors loading.
#3930 opened May 18, 2026 by copybara-service Bot Loading…
@abhinavclemson
1
NNX: load_parameters_path adapter for Linen<->NNX checkpoints gemini-review
#3929 opened May 18, 2026 by ecnal-cienet Collaborator Loading…
4 tasks done
7
Gemma 2 Local/Global KV Cache Split
#3926 opened May 17, 2026 by learning-to-play Collaborator Loading…
1
Gemma and Gemma 2 Native Serving Registration
#3925 opened May 17, 2026 by learning-to-play Collaborator Loading…
1
Updates 'full' remat policy to use nothing_saveable
#3924 opened May 17, 2026 by learning-to-play Collaborator Loading…
1
Clean up YAML files: lowercase booleans, trailing whitespace, and yamllint integration gemini-review
#3923 opened May 15, 2026 by igorts-git Collaborator Loading…
4 tasks done
5
[ROCm] optimizations to reduce temp size prior to training starts
#3922 opened May 15, 2026 by cj401-amd Loading…
6
Nicogrande/add gdn support
#3917 opened May 14, 2026 by NicoGrande Collaborator • Draft
4 tasks
1
ProTip! Follow long discussions with comments:>50.