Skip to content

Pull requests: NVIDIA-NeMo/RL

New pull request New
102 Open 1,053 Closed
102 Open 1,053 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

docs: Add notes for FP8 recipe in docs/fp8.md CI:docs Run doctest documentation Improvements or additions to documentation
#1829 opened Jan 26, 2026 by guyueh1 Loading…
4 tasks
@guyueh1
3
feat: add lora config for dpo dtensor backend CI:L1 Run doctests, unit tests, and functional tests
#1826 opened Jan 26, 2026 by RayenTian Loading…
4 tasks
3
feat: add FT launcher config and resiliency dependency [1/4] CI:docs Run doctest documentation Improvements or additions to documentation
#1824 opened Jan 23, 2026 by yashaswikarnati Loading…
4 tasks
15
ci: introduce renovate to deal with bumping our dependencies CI Relating to CI
#1823 opened Jan 23, 2026 by terrykong • Draft
4 tasks
5
ci: Allow repo to self publish docs CI Relating to CI
#1821 opened Jan 23, 2026 by chtruong814 Loading…
4 tasks
7
perf: Update cudnn to 9.14 CI:L2 Run doctests, unit tests, functional tests, and convergence tests
#1820 opened Jan 23, 2026 by guyueh1 Loading…
4 tasks
1
@guyueh1
4
fix: fix statistic of probs_ratio_clamped_min/max CI:L1 Run doctests, unit tests, and functional tests
#1818 opened Jan 23, 2026 by yuki-97 Loading…
6
chore: add assert for dtensor v2 cpu offload
#1817 opened Jan 23, 2026 by yuki-97 Loading…
3
fix: Unify custom model logits extraction across all inference methods CI:L2 Run doctests, unit tests, functional tests, and convergence tests
#1815 opened Jan 23, 2026 by zpqiu Loading…
4 tasks
11
feat: Implement ProRLv2 recipe
#1809 opened Jan 22, 2026 by hijkzzz Loading…
28
feat: unify nemogym dataset
#1807 opened Jan 22, 2026 by yuki-97 • Draft
chore: cuda13 support CI:L2 Run doctests, unit tests, functional tests, and convergence tests
#1803 opened Jan 21, 2026 by guyueh1 Loading…
4 tasks
7
feat: Timer for the data sharding and job submission CI:L2 Run doctests, unit tests, functional tests, and convergence tests
#1802 opened Jan 21, 2026 by guyueh1 Loading…
4 tasks
@guyueh1
16
feat: Support lora in dtensor grpo workflow by merging weight CI:L1 Run doctests, unit tests, and functional tests
#1797 opened Jan 20, 2026 by RayenTian • Draft
5
Bxyu/gym dynamic sampling
#1793 opened Jan 18, 2026 by bxyu-nvidia • Draft
4 tasks
Feat: Megatron LoRA GRPO sync colocated [1/3]
#1790 opened Jan 17, 2026 by vadam5 • Draft
4 tasks
3
[Draft] Update Megatron submodule pins
#1787 opened Jan 16, 2026 by yaoyu-33 • Draft
4
feat: add speculative decoding during post-training
#1785 opened Jan 15, 2026 by isomap Loading…
2 of 4 tasks
4
feat: NeMo Gym GRPO on-policy fix params; Per-agent group-level rewards CI:L1 Run doctests, unit tests, and functional tests
#1779 opened Jan 15, 2026 by bxyu-nvidia Loading…
4 tasks
10
[don't merge] split train and val dataset in preference dataset CI:L1 Run doctests, unit tests, and functional tests documentation Improvements or additions to documentation
#1763 opened Jan 13, 2026 by yuki-97 • Draft
1
19
[docs] Document Gym + RL integration design documentation Improvements or additions to documentation
#1762 opened Jan 12, 2026 by ananthsub • Draft
1 of 4 tasks
feat: refactor train utilities for dtensor policy v2
#1757 opened Jan 10, 2026 by hemildesai • Draft
4 tasks
1
feat: Adding quantization aware training support with Model-Optimizer
#1756 opened Jan 10, 2026 by mxinO • Draft
4 tasks
2
feat: Support lora in dtensor grpo workflow[3/3]: async vllm CI:L1 Run doctests, unit tests, and functional tests
#1752 opened Jan 9, 2026 by RayenTian Loading…
7 tasks
3
feat: Support lora in dtensor grpo workflow[2/3]: sync and non-colocated setup CI:L1 Run doctests, unit tests, and functional tests
#1751 opened Jan 9, 2026 by RayenTian Loading…
4 tasks
13
ProTip! Updated in the last three days: updated:>2026-01-24.