-
Notifications
You must be signed in to change notification settings - Fork 269
Pull requests: google/tunix
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix corner case exception for kl computation
#1344
opened Apr 1, 2026 by
copybara-service
bot
Loading…
Move RLOO advantage estimator from GRPOLearner to AgenticGRPOLearner.
#1340
opened Apr 1, 2026 by
copybara-service
bot
Loading…
Fix LoRA merge for sharded Gemma 3 checkpoints and HF parameter mapping
#1336
opened Apr 1, 2026 by
ayehninnkhine
Loading…
Simplify tunix/oss file transformations in copy.bara.sky.
#1335
opened Apr 1, 2026 by
copybara-service
bot
Loading…
[Tunix] Add dual-clip on pg_loss in agentic GRPO learner and compute kl_loss before applying kl penalty.
#1334
opened Mar 31, 2026 by
copybara-service
bot
Loading…
Introducing a compatibility layer to convert v0
CheckpointManagerOptions to v1 policies.
#1330
opened Mar 31, 2026 by
copybara-service
bot
Loading…
[tunix/sft] Add
shard_input_data flag to PeftTrainer
#1325
opened Mar 28, 2026 by
copybara-service
bot
Loading…
Fix Qwen2 KV-cache dtype mismatch in cached decoding
#1322
opened Mar 27, 2026 by
skwh54
Loading…
6 tasks done
improve rl_cluster init corner case handling and test coverage
#1311
opened Mar 26, 2026 by
copybara-service
bot
Loading…
fix file path for window, : is reserved name on window
#1307
opened Mar 26, 2026 by
HugoTian
Loading…
security: replace eval() with safe AST math evaluator in calculate reward
#1279
opened Mar 23, 2026 by
RaymondSeven
Loading…
Previous Next
ProTip!
Find all pull requests that aren't related to any open issues with -linked:issue.