Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
183 Open 2,592 Closed
183 Open 2,592 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 TE NVFP4 quantization config without RHT
#2987 opened Jan 21, 2026 by hx89 Loading…
4 tasks
Force layout on Q for MLA.
#2986 opened Jan 21, 2026 by copybara-service bot Loading…
1
Print logic axes
#2985 opened Jan 21, 2026 by Shuang-cnt • Draft
4 tasks done
1
[src/MaxText/inference/offline_engine.py] Make OfflineEngine a fluent interface
#2983 opened Jan 21, 2026 by SamuelMarks Loading…
4 tasks done
1
[src/MaxText/input_pipeline/_tfds_data_processing.py] Builder Pattern for TFDS Preprocessing Pipeline: preprocessing_pipeline
#2982 opened Jan 21, 2026 by SamuelMarks Loading…
4 tasks done
1
[src/MaxText/input_pipeline/_hf_data_processing.py] Refactor to fluent interface for constructing Grain pipelines (introduce GrainPipelineBuilder); increase type-safety
#2981 opened Jan 21, 2026 by SamuelMarks Loading…
4 tasks done
1
NNX migration: NNX utils and unit tests
#2980 opened Jan 21, 2026 by xibinliu • Draft
4 tasks done
1
Add flops calculation for DeepSeek v3.2
#2979 opened Jan 21, 2026 by RissyRan Loading…
4 tasks done
@shuningjin @gagika
1
Adding chat template to vllm decode.
#2978 opened Jan 20, 2026 by NicoGrande Loading…
4 tasks done
1
Tyagiva diagon upgrade
#2976 opened Jan 20, 2026 by tyagivaji Loading…
4 tasks done
4
Record maxtext & post-training requirements commit ids to CI workflow
#2975 opened Jan 20, 2026 by SurbhiJainUSC • Draft
4 tasks done
1
checkpoint utility: shard checkpoint, monitor peak
#2974 opened Jan 20, 2026 by shuningjin Loading…
4 tasks done
@RissyRan @hengtaoguo @NicoGrande
1
Add Qwen3-Next to checkpoint util
#2973 opened Jan 20, 2026 by Rohan-Bierneni Loading…
4 tasks done
1
Add caching for GatedDeltaNetCache (linear attention models)
#2971 opened Jan 20, 2026 by Rohan-Bierneni Loading…
4 tasks done
1
Support system role in hf data processing gemini-review
#2970 opened Jan 20, 2026 by ChingTsai Loading…
4 tasks done
@ChingTsai
5
Enable compute offloading in MaxText
#2969 opened Jan 20, 2026 by zhenying-liu • Draft
4 tasks
Implement QK-Clip (Muon-Clip) for MLA attention
#2968 opened Jan 20, 2026 by gagika • Draft
4 tasks
[src/MaxText/utils/ckpt_conversion] Split into one file per ckpt_conversion following strategy pattern
#2967 opened Jan 20, 2026 by SamuelMarks Loading…
4 tasks done
1
Chain of Responsibility for Logits Processing
#2966 opened Jan 20, 2026 by SamuelMarks Loading…
4 tasks done
1
NNX migration preparation: pure_nnx flag and init_state_fn
#2965 opened Jan 19, 2026 by xibinliu Loading…
4 tasks done
1
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…
4 tasks done
8
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 Loading…
4 tasks done
6
Improve formatting and clarify MFU comparison caveat in performance metrics docs.
#2959 opened Jan 16, 2026 by copybara-service bot Loading…
@shralex
ProTip! no:milestone will show everything without a milestone.