Skip to content

Pull requests: NovaSky-AI/SkyRL

New pull request New
68 Open 726 Closed
68 Open 726 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 LoRA sharding compatibility with JAX 0.9+
#1040 opened Feb 7, 2026 by jaredquincy Loading…
2 of 3 tasks
3
[train] Support logprobs, fix generation config defaults and add more generation tests for the new HTTP inference pathway
#1038 opened Feb 6, 2026 by SumanthRH Loading…
6
Implement DPPO Algorithm
#1037 opened Feb 6, 2026 by jonahsamost • Draft
wip - glm flash 4.7 support for megatron
#1033 opened Feb 6, 2026 by erictang000 • Draft
[tx] Implement stacked weights tx
#1018 opened Feb 4, 2026 by pcmoritz Loading…
9
[WIP] Copy tinker to skyrl-train to eliminate cross-repo dependencies
#1012 opened Feb 2, 2026 by tyler-griggs • Draft
4 tasks
[tx] General implementation of trainable Hyper Connections tx
#1008 opened Feb 2, 2026 by tanmaysachan Loading…
2 tasks done
8
Remove LR scheduler for workers
#1002 opened Feb 1, 2026 by jonahsamost Loading…
2
[tx] Per-layer gradient checkpointing with stacked decoder layers tx
#996 opened Jan 30, 2026 by raulchen Loading…
44
[tx] Add GLM4.7 tx
#989 opened Jan 29, 2026 by tanmaysachan Loading…
3 tasks done
8
Add Tinker API server for supervised fine-tuning with SkyRL backend
#986 opened Jan 29, 2026 by tyler-griggs • Draft
4 tasks
[tx] Support LoRA in the unembedding layer, redux tx
#984 opened Jan 28, 2026 by pcmoritz Loading…
6
fix: rope_scaling for vLLM and HuggingFace context extension
#976 opened Jan 27, 2026 by dzorlu Loading…
1 task
[tx] Add max_micro_batches to limit batch size for forward/forward_backward requests tx
#968 opened Jan 26, 2026 by raulchen Loading…
4
[WIP] [tx] Make it easy to run on a multi-node Ray cluster tx
#955 opened Jan 25, 2026 by Future-Outlier Loading…
7
Separate infrastructure logs from training progress
#953 opened Jan 25, 2026 by tyler-griggs Loading…
2 of 3 tasks
4
Add SKYRL_LOG_LEVEL env var and suppress verbose Ray logs by default
#947 opened Jan 25, 2026 by tyler-griggs Loading…
1 of 2 tasks
2
Stage 5: Add TinkerTrainingAdapter for forward_backward/optim_step
#938 opened Jan 24, 2026 by tyler-griggs Loading…
7
[skyrl-train] Fix loss reduction by moving normalization to the advantage computation
#925 opened Jan 23, 2026 by justinvyu Loading…
5
Add loss_fn parameterization to forward_backward
#924 opened Jan 23, 2026 by tyler-griggs • Draft
3
DSPy integration
#921 opened Jan 22, 2026 by Harryllh Loading…
[skyrl-train] Add GDPO Support to PPO Utils
#897 opened Jan 19, 2026 by devpatelio Loading…
2 tasks
3
[tx] Implement cutlass kernel for ragged_dot with group_offset tx
#896 opened Jan 19, 2026 by pcmoritz Loading…
4
[skyrl-train] Update Megatron + LoRA to use PEFT only bridge
#885 opened Jan 15, 2026 by erictang000 • Draft
1 of 3 tasks
[algorithm] Sampling mask from Inference Engine Applies to Trainer
#883 opened Jan 15, 2026 by devpatelio Loading…
9
ProTip! What’s not been updated in a month: updated:<2026-01-07.