Skip to content

Pull requests: NVIDIA-NeMo/RL

New pull request New
56 Open 418 Closed
56 Open 418 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

feat: save checkpoint before timeout to avoid 4-hour runtime limit
#734 opened Jul 24, 2025 by wedu-nvidia Loading…
4 tasks
cp: docs: remove doc duplicated (721) into r0.3.0 cherry-pick documentation Improvements or additions to documentation Run CICD Set to run CI (unset + set to rerun)
#733 opened Jul 23, 2025 by chtruong814 • Queued
feat: Enable simulated user for multi-turn GRPO [new]
#732 opened Jul 23, 2025 by jialei777 Loading…
4 tasks
doc: Update Frontpage README with new features
#731 opened Jul 23, 2025 by SahilJain314 Loading…
1
fix: guard DPO against dynamic batching and sequence packing r0.3.0 Release r0.3.0
#730 opened Jul 23, 2025 by ashors1 • Queued
4 tasks
2
docs: Added docs for sequence packing and dynamic batching documentation Improvements or additions to documentation r0.3.0 Release r0.3.0
#729 opened Jul 23, 2025 by SahilJain314 Loading…
3
docs: Update docs to include submodule instructions documentation Improvements or additions to documentation r0.3.0 Release r0.3.0
#725 opened Jul 23, 2025 by yfw Loading…
4 tasks
feat: Overlong filtering for GRPO documentation Improvements or additions to documentation
#724 opened Jul 23, 2025 by jubick1337 Loading…
4 tasks
feat(logger.py): support swanlab documentation Improvements or additions to documentation external
#716 opened Jul 23, 2025 by tpoisonooo Loading…
4 tasks
3
fix: Use the conditional temperature scaling in get_logprobs as well r0.3.0 Release r0.3.0
#714 opened Jul 22, 2025 by parthchadha Loading…
4 tasks
3
test: Add Megatron tests
#713 opened Jul 22, 2025 by ashors1 • Draft
4 tasks
feat: SFT support for multimodal training (VLM)
#712 opened Jul 22, 2025 by rohitrango • Draft
1 of 3 tasks
fix: OOM with some GRPO configs
#709 opened Jul 22, 2025 by ahmadki Loading…
4 tasks
4
feat: added save_to_parquet feature to eval
#708 opened Jul 22, 2025 by shaoxiongduan Loading…
1 of 4 tasks
13
chore: remove old fsdp1 args everywhere
#707 opened Jul 22, 2025 by terrykong Loading…
4 tasks
feat: enable toggling between thinking and non-thinking for evaluation. documentation Improvements or additions to documentation
#702 opened Jul 21, 2025 by xxman-google Loading…
4 tasks done
8
fix: remove tie weight check CI:L1 Run doctests, unit tests, and functional tests
#700 opened Jul 21, 2025 by RayenTian Loading…
1
7
ci: Enforce code coverage CI:L0 Run doctests and unit tests CI Relating to CI
#694 opened Jul 18, 2025 by chtruong814 Loading…
4 tasks
1
Fix data len
#690 opened Jul 18, 2025 by joyang-nv • Draft
4 tasks
feat: refit metadata optimization
#686 opened Jul 17, 2025 by ZhiyuLi-Nvidia Loading…
3 of 5 tasks
11
perf(model_utils.py): avoiding softmax deepcopy external
#683 opened Jul 17, 2025 by tpoisonooo Loading…
3
feat(run_grpo_math.py): support local data dir
#677 opened Jul 16, 2025 by tpoisonooo Loading…
4 tasks
2
chore: switch from mypy to pyrefly CI Relating to CI
#675 opened Jul 16, 2025 by terrykong Loading…
9
feat: preference datasets algorithm documentation Improvements or additions to documentation enhancement New feature or request training Training related
#673 opened Jul 15, 2025 by jveronvialard • Draft
4 tasks done
52
feat: v0 VLM support + GRPO pipeline CI:L1 Run doctests, unit tests, and functional tests
#655 opened Jul 11, 2025 by rohitrango Loading…
1 of 4 tasks
@rohitrango
20
ProTip! What’s not been updated in a month: updated:<2025-06-23.