Skip to content

Pull requests: NovaSky-AI/SkyRL

New pull request New
138 Open 1,179 Closed
138 Open 1,179 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

[ci] H100 CI run_train_gpu_ci run_train_megatron_gpu_ci_models run_train_megatron_gpu_ci
#1679 opened May 16, 2026 by hao-aaron Collaborator Loading…
6
vllm metrics docs
#1662 opened May 13, 2026 by hao-aaron Collaborator Loading…
10
[vllm] test out ray v2 executor run_train_gpu_ci run_train_megatron_gpu_ci_models run_train_megatron_gpu_ci run_train_old_inference_gpu_ci
#1660 opened May 13, 2026 by erictang000 Collaborator Loading…
3
[tx] add retries for getting random port used for JAX coordinator address
#1652 opened May 12, 2026 by andrewsykim Contributor Loading…
2
Avoid blocking Ray registry lookups for policy loss resolution
#1644 opened May 10, 2026 by taivu1998 Loading…
1
[tinker] Support additional LoRA config parameters
#1643 opened May 10, 2026 by taivu1998 Loading…
1
8
Fix multi-turn generator token budget accounting
#1642 opened May 10, 2026 by taivu1998 Loading…
1
3
Add opt-in handling for failed SkyRLGym rollouts
#1641 opened May 10, 2026 by taivu1998 Loading…
1
4
Stabilize Qwen3 MoE LoRA mixed-scale test
#1640 opened May 10, 2026 by taivu1998 Loading…
1
Remove legacy skyrl-train and skyrl-tx folders
#1639 opened May 10, 2026 by taivu1998 Loading…
1
2
[skyrl] Pass through profiler_config to vLLM engine
#1622 opened May 5, 2026 by pcmoritz Collaborator Loading…
3
[worktrial] Taste reward shaping
#1618 opened May 4, 2026 by alliegu-fleet Loading…
5
Fix prompt batch alignment validation for synchronous PPO
#1615 opened May 3, 2026 by taivu1998 Loading…
1
3
[train] Support streaming mini-batch (non-blocking async training)
#1607 opened May 1, 2026 by rishithayenumula Loading…
1
4
[train] Use custom wheel for vllm-router for /chat/completions fix run_train_gpu_ci run_train_megatron_gpu_ci
#1601 opened Apr 30, 2026 by SumanthRH Member Loading…
6
Fix GPU assignment for Slurm-launched Ray clusters
#1592 opened Apr 29, 2026 by agolajko Contributor Loading…
1
4
Fix AssertionError during eval when val set size is not divisible by train_batch_size
#1589 opened Apr 29, 2026 by rishithayenumula Loading…
1
@CharlieFRuan
8
[models] add gemma4 training script
#1576 opened Apr 27, 2026 by erictang000 Collaborator • Draft
[train] Fix rollout metrics for step-wise and custom generators (sync / fully async)
#1556 opened Apr 22, 2026 by CharlieFRuan Member • Draft
1 of 3 tasks
6
[WIP] Add changes needed for FP8 megatron training
#1543 opened Apr 21, 2026 by pcmoritz Collaborator Loading…
9
[train][step-wise] Three correctness/efficiency fixes for step-wise training
#1539 opened Apr 20, 2026 by CharlieFRuan Member Loading…
4 tasks done
2
[skyrl] Preserve staged forward_backward loss_fn_outputs across DP ranks
#1534 opened Apr 19, 2026 by taivu1998 Loading…
1
9
Modify SkyRL Generator to Append Router Indices in Multi-Turn
#1530 opened Apr 17, 2026 by devpatelio Collaborator Loading…
5
Add FoldGRPO advantage estimator and process_rewards pipeline
#1514 opened Apr 15, 2026 by sumi-fleet-hub Loading…
@CharlieFRuan
6
SFT loss aggregation consistent with RL path
#1513 opened Apr 15, 2026 by agolajko Contributor Loading…
@SumanthRH @erictang000
11
ProTip! Follow long discussions with comments:>50.