Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
178 Open 2,575 Closed
178 Open 2,575 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

Pydantic class TrainContext for train functions (replacing old 11-arity function)
#2964 opened Jan 17, 2026 by SamuelMarks Loading…
4 tasks done
1
doc: add checkpoint conversion in tutorials
#2963 opened Jan 16, 2026 by charlesli640 Loading…
1 task
1
Move tests/forward_pass_logit_checker.py to tests/utils/forward_pass_logit_checker.py gemini-review
#2962 opened Jan 16, 2026 by SurbhiJainUSC Loading…
4 tasks done
4
feat: Add HuggingFace to MaxText converter for Gemma 3
#2961 opened Jan 16, 2026 by yasxai Loading…
1
feat: update knowledge distillation tutorial for using vllm with Qwen model
#2960 opened Jan 16, 2026 by RexBearIU • Draft
4 tasks
Improve formatting and clarify MFU comparison caveat in performance metrics docs.
#2959 opened Jan 16, 2026 by copybara-service bot Loading…
@shralex
Adjusted the batch size to enable llama2-70b to run on v5p-128
#2957 opened Jan 16, 2026 by hsuan-lun-chiang Loading…
4 tasks done
1
Move src/MaxText/distillation to src/maxtext/trainers/post_train/distillation gemini-review
#2956 opened Jan 15, 2026 by SurbhiJainUSC Loading…
4 tasks done
@shralex @gagika
12
Support attention data parallelism
#2955 opened Jan 15, 2026 by khatwanimohit Loading…
4 tasks done
5
Add option to skip first token during logits comparison
#2953 opened Jan 15, 2026 by gagika Loading…
4 tasks done
@RissyRan @shuningjin
3
Add Qwen3 Audio Encoder
#2952 opened Jan 15, 2026 by aireenmei Loading…
4 tasks done
10
Move integration tests to tests/integration and unit tests to tests/unit gemini-review
#2949 opened Jan 14, 2026 by SurbhiJainUSC Loading…
4 tasks done
@shralex
4
Set local_devices from CUDA_VISIBLE_DEVICES in initialize_jax_for_gpu
#2945 opened Jan 14, 2026 by gabeweisz Loading…
4 tasks done
1
replace uses of pltpu.repeat with jnp.tile in pallas.
#2943 opened Jan 14, 2026 by copybara-service bot Loading…
1
replace uses of pltpu.repeat with jnp.tile in pallas.
#2939 opened Jan 13, 2026 by copybara-service bot Loading…
1
Feat/migrate maxengine to nnx
#2937 opened Jan 13, 2026 by ecnal-cienet • Draft
4 tasks done
DeepSeek3.2: Onboard sparse attention
#2933 opened Jan 13, 2026 by RissyRan Loading…
4 tasks done
@RissyRan @shuningjin @gagika
24
[DRAFT] Insert gathers draft Draft PR
#2932 opened Jan 13, 2026 by NuojCheng • Draft
4 tasks
1
Trigger unit tests for docker images upload
#2924 opened Jan 9, 2026 by xibinliu Loading…
4 tasks done
1
[Diloco] Diloco trainer
#2920 opened Jan 8, 2026 by khatwanimohit • Draft
4 tasks done
13
Insert all gather per repeats for pipeline parallelism draft Draft PR
#2919 opened Jan 7, 2026 by NuojCheng • Draft
4 tasks
1
Happy New Year | Copyright 2023–2026 Google LLC
#2916 opened Jan 7, 2026 by SamuelMarks Loading…
4 tasks done
1
[src/MaxText/pyconfig.py] New initialize_pydantic_with_maxtext_config function ; [src/MaxText/examples/rl_llama3_demo.ipynb] Use new function
#2915 opened Jan 7, 2026 by SamuelMarks Loading…
4 tasks done
1
[Diloco] add diloco related utils
#2913 opened Jan 6, 2026 by khatwanimohit Loading…
4 tasks done
1
Add Packing Support for Context Parallelism (Ring Attention)
#2906 opened Dec 31, 2025 by kocchop Loading…
4 tasks done
14
ProTip! Exclude everything labeled bug with -label:bug.