Skip to content

Pull requests: google/tunix

New pull request New
81 Open 809 Closed
81 Open 809 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

extend support for gemma1.1 in automodel
#970 opened Jan 13, 2026 by copybara-service bot Loading…
Add gcsfs.
#969 opened Jan 12, 2026 by copybara-service bot Loading…
Fix PeftTrainer input sharding for raw inputs and add regression test
#967 opened Jan 12, 2026 by jayshrivastava0 Loading…
6 tasks done
1
[Not ready for review] Memory Optimizations on Gemma2 2B
#966 opened Jan 12, 2026 by Green0-0 Loading…
1
fix: Add truncation support to pad_to_length for consistent prompt shapes
#964 opened Jan 11, 2026 by Cascoopman Loading…
Fix batch_size=1 token_buffer truncation in sampler
#963 opened Jan 11, 2026 by Cascoopman Loading…
1
feat(agentic): Add tool calling support for GRPO training
#961 opened Jan 11, 2026 by Cascoopman Loading…
Fix #802: Add google-cloud-storage and gcsfs to dependencies
#958 opened Jan 9, 2026 by Aaryan-549 Loading…
3 of 5 tasks
1
Add overlong reward shaping for DAPO.
#947 opened Jan 6, 2026 by copybara-service bot Loading…
Add CRD distillation strategy and tests
#933 opened Dec 22, 2025 by surajyadav-research Loading…
1
[Tunix] Update dataset initialization in nightly regression script and add to CI.
#932 opened Dec 22, 2025 by copybara-service bot Loading…
fix(utils): correct positive integer validation
#930 opened Dec 21, 2025 by parthmaheshwari-hub Loading…
1
Refactor tokenizer adapter to support special EOS token
#927 opened Dec 21, 2025 by kamalkraj Loading…
6 tasks done
[Tunix] Remove intermediate activation sharding in FeedForward blocks.
#921 opened Dec 18, 2025 by copybara-service bot Loading…
[Tunix] Add agentic mode support to grpo_demo_llama3_qwen2.py.
#920 opened Dec 18, 2025 by copybara-service bot Loading…
[Tunix] Reorder TPU nightly regression workflow steps.
#906 opened Dec 17, 2025 by copybara-service bot Loading…
Fix AttributeError in is_positive_integer on Python 3.11
#903 opened Dec 17, 2025 by bug39 Loading…
5
fix broken test.
#898 opened Dec 16, 2025 by copybara-service bot Loading…
Upgrade GitHub Actions to latest versions
#895 opened Dec 16, 2025 by salmanmkc Loading…
4
Upgrade GitHub Actions for Node 24 compatibility
#894 opened Dec 16, 2025 by salmanmkc Loading…
7
Fdpo
#892 opened Dec 16, 2025 by xinzhuo20 Loading…
3
Add metrics support to CheckpointManager and PeftTrainer
#885 opened Dec 14, 2025 by kamalkraj Loading…
6 tasks done
10
Add Qwen3 235B-A22B model config
#871 opened Dec 10, 2025 by yhtang Loading…
6 tasks done
1
Update sharding for activations to include 'dp' axis. Validated on internal notebook, after enabling (2, 1, 4) for (dp, fsdp, tp), do see weights replicated on dp axis.
#849 opened Dec 5, 2025 by copybara-service bot Loading…
Explicitly cast input_ids to jnp.int32
#838 opened Dec 4, 2025 by copybara-service bot Loading…
ProTip! Exclude everything labeled bug with -label:bug.