Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
226 Open 4,024 Closed
226 Open 4,024 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

Bump the pip group across 2 directories with 4 updates dependencies Pull requests that update a dependency file python Pull requests that update python code
#4527 opened Jul 17, 2026 by dependabot Bot Loading…
Add forward_pass_compile_test script
#4526 opened Jul 17, 2026 by snehalv2002 Collaborator Loading…
4 tasks done
2
Fix NNX MTP zero-loss, qwix MTP crash, and DeepSeek scan-axis sharding corruption
#4525 opened Jul 17, 2026 by ecnal-cienet Collaborator • Draft
4 tasks done
2
[WIP] Weight Sync
#4524 opened Jul 17, 2026 by YixuanWang-99 Collaborator Loading…
4 tasks
1
Split logical rule for eval from for train
#4523 opened Jul 17, 2026 by NuojCheng Collaborator • Draft
4 tasks done
2
Decouple vision encoder selection from model name
#4522 opened Jul 17, 2026 by hengtaoguo Collaborator Loading…
4 tasks done
3
Fix ring ragged unsort gradient and clamp local group size.
#4521 opened Jul 17, 2026 by copybara-service Bot Loading…
@RissyRan
1
Extend configuration and improve data loader/checkpointing exception resiliency for elastic training.
#4520 opened Jul 17, 2026 by copybara-service Bot Loading…
@abhinavclemson
1
fix: resolve vLLM training gibberish outputs for Gemma 3/4 scanned layers
#4519 opened Jul 17, 2026 by mesakhcienet Collaborator Loading…
4 tasks done
Fix/routed moe test
#4518 opened Jul 17, 2026 by mesakhcienet Collaborator • Draft
4 tasks
1
Add Qwen3-VL-30B-A3B
#4517 opened Jul 17, 2026 by lydhr Collaborator Loading…
4 tasks done
1
Optional block-diffusion support for Qwen3 (attention + masked-diffusion SFT + on-policy distillation + RL rollout)
#4516 opened Jul 17, 2026 by ethannnnnn Loading…
1
fix(test): brittle string assertions in decode sampling tests
#4515 opened Jul 17, 2026 by mesakhcienet Collaborator Loading…
4 tasks done
2
Clarify HBM memory estimation log
#4514 opened Jul 17, 2026 by Shuwen-Fang Collaborator Loading…
4 tasks done
3
Support ragged_sort_use_single_sparsecore flag to restrict token sorting to 1 SparseCore
#4513 opened Jul 16, 2026 by Shuwen-Fang Collaborator Loading…
4 tasks done
4
test: add numerical reference coverage for the Qwen3.5 text decoder c…
#4512 opened Jul 16, 2026 by chloechiaw Loading…
2 of 4 tasks
Update gmm2 fork
#4511 opened Jul 16, 2026 by shuningjin Collaborator • Draft
4 tasks
fix: Gemma3-4b E2E script
#4509 opened Jul 16, 2026 by chiajunglien Contributor • Draft
4 tasks done
9
add nnx_scan util module pull ready
#4507 opened Jul 16, 2026 by aireenmei Collaborator Loading…
4 tasks done
1
Fix QLoRA and Gemma4-small decoder issues for NNX
#4504 opened Jul 16, 2026 by hsuan-lun-chiang Collaborator • Draft
4 tasks done
1
feat(nnx): support native Flax NNX PEFT/LoRA training loop
#4501 opened Jul 16, 2026 by RexBearIU Collaborator Loading…
4 tasks done
4
fix: OOM issue
#4500 opened Jul 16, 2026 by chiajunglien Contributor • Draft
4 tasks
Add chunked MoE support to MaxText. pull ready
#4499 opened Jul 16, 2026 by copybara-service Bot Loading…
2 tasks done
1
Migrate Tunix's SFT Checkpoint Manager usage of Orbax V0 Checkpoint Manager to V1 Checkpointer.
#4498 opened Jul 15, 2026 by copybara-service Bot Loading…
1
Dsv4 nnx support
#4497 opened Jul 15, 2026 by snehalv2002 Collaborator Loading…
3 of 4 tasks
2
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.