-
Notifications
You must be signed in to change notification settings - Fork 461
Pull requests: AI-Hypercomputer/maxtext
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Fix duplicate peak learning rate in warmup schedule
#3095
opened Feb 5, 2026 by
ChingTsai
Loading…
4 tasks done
Add support for overriding model architecture in Hugging Face conversion
#3094
opened Feb 5, 2026 by
gagika
Loading…
Fix the default value of tokenizer_path
pull ready
#3092
opened Feb 5, 2026 by
zxhe-sean
Loading…
4 tasks done
Update paths and types in order to run train_distill.
gemini-review
pull ready
#3091
opened Feb 5, 2026 by
entrpn
Loading…
4 tasks done
Test for generate_param_only_checkpoint_test
#3090
opened Feb 5, 2026 by
hengtaoguo
•
Draft
4 tasks done
Integrate DeepSeek Sparse Attention with Tokamax Flash Attention
gemini-review
#3087
opened Feb 4, 2026 by
RissyRan
Loading…
4 tasks done
CI test: divide to 2 worker groups for more UT
#3086
opened Feb 4, 2026 by
charlesli640
•
Draft
4 tasks done
Dump activation shardings
draft
Draft PR
#3080
opened Feb 4, 2026 by
charlesli640
•
Draft
4 tasks done
Roll forward after fix: https://github.com/AI-Hypercomputer/maxtext/pull/3050
#3079
opened Feb 4, 2026 by
copybara-service
bot
Loading…
[Do Not Merge] Optimizations on Qwen3-Next GatedDeltaNet w/ Kernel & XProf Agent
#3077
opened Feb 4, 2026 by
Rohan-Bierneni
Loading…
4 tasks done
Deepseek sharding for vLLM and MLA kernel plumbing
#3072
opened Feb 3, 2026 by
khatwanimohit
•
Draft
4 tasks done
Remove DPO (Direct Preference Optimization) feature
#3064
opened Feb 2, 2026 by
ecnal-cienet
Loading…
4 tasks done
[MaxEngine] Fix TypeError in prefill() during batched inference
#3063
opened Feb 2, 2026 by
jaisong123
Loading…
Previous Next
ProTip!
Adding no:label will show everything without a label.