Skip to content

Pull requests: NovaSky-AI/SkyRL

New pull request New
125 Open 1,126 Closed
125 Open 1,126 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

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
[feat] Multi-LoRA serving for RemoteInferenceClient
#1579 opened Apr 28, 2026 by hao-aaron Collaborator Loading…
3
[models] add gemma4 training script
#1576 opened Apr 27, 2026 by erictang000 Collaborator • Draft
Widen transformers for v5.6 and vllm==0.19.1
#1561 opened Apr 22, 2026 by jamesbraza Loading…
1
@erictang000
6
[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
fix(docker): optimize Dockerfile.megatron to reduce image size by 1.36 GB run_train_megatron_gpu_ci
#1499 opened Apr 11, 2026 by dinhxuanvu Loading…
11
Add ppo as alias for dual_clip policy loss type
#1481 opened Apr 8, 2026 by j316chuck • Draft
feat: add max_tokens_per_microbatch config for token-based micro-batching
#1477 opened Apr 8, 2026 by erictang000 Collaborator Loading…
5
feat: native Atropos-SHM integration and modular ingestion layer
#1473 opened Apr 7, 2026 by RUFFY-369 Loading…
10
[train] Enable expandable_segments to reduce GPU memory fragmentation run_train_gpu_ci
#1470 opened Apr 7, 2026 by CharlieFRuan Member • Draft
5 tasks done
1
7
[tinker] Support prompt_logprobs in SkyRLTrainBackend sample() path
#1461 opened Apr 6, 2026 by pbokc Contributor Loading…
21
[tinker] Support KL loss in SkyRLTrainBackend
#1460 opened Apr 5, 2026 by pbokc Contributor Loading…
9
feat: LLM-synthesized hints for failed trajectories
#1456 opened Apr 4, 2026 by dzorlu Loading…
4 tasks
6
[skyrl-train] feat: add native GMPO policy loss with validation and tests
#1449 opened Apr 2, 2026 by taivu1998 Loading…
@erictang000
3
Fix event-loop blocking in one-step-off async save/export paths
#1446 opened Apr 2, 2026 by taivu1998 Loading…
1
@erictang000
3
Change default KL estimator from k3 to k2 for loss-based KL
#1445 opened Apr 2, 2026 by taivu1998 Loading…
@erictang000
3
[SkyRLGymGenerator] Respect remaining context budget in agent_loop
#1441 opened Apr 2, 2026 by taivu1998 Loading…
4
[skyrl-train] Flip grpo_norm_by_std default to false
#1443 opened Apr 2, 2026 by taivu1998 Loading…
1
ProTip! Follow long discussions with comments:>50.