Skip to content

Pull requests: NVIDIA-NeMo/RL

New pull request New
61 Open 567 Closed
61 Open 567 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: add config_tools.py and refactor configs
#1024 opened Aug 29, 2025 by terrykong • Draft
feat: Implement safetensors checkpointing format support using nemo-automodel
#1023 opened Aug 29, 2025 by ffrujeri • Draft
4 tasks done
2
fix: ray.sub will exit early if any srun fails to launch
#1022 opened Aug 29, 2025 by terrykong Loading…
1
chore: flush to stdout when print logging during GRPO
#1021 opened Aug 29, 2025 by pjin-nvidia Loading…
4 tasks
fix: fix convergence issue by adding use_inductor=False in vllm compi…
#1014 opened Aug 28, 2025 by ZhiyuLi-Nvidia Loading…
1 of 4 tasks
4
feat: add on policy distillation algorithm algorithm
#1006 opened Aug 28, 2025 by zpqiu • Draft
4 tasks
1
feat: Fix nsight profiling file sync for multi-node jobs documentation Improvements or additions to documentation
#1001 opened Aug 27, 2025 by guyueh1 • Queued
4 tasks
1
6
ci: Fix build and test publish wheel CI Relating to CI
#995 opened Aug 27, 2025 by chtruong814 Loading…
4 tasks
14
draft: feat: fused loss and logit to logprob conversion
#994 opened Aug 27, 2025 by jiemingz Loading…
4 tasks
2
feat: Integrate vlm changes between DTensorPolicyWorker V1 and V2.
#982 opened Aug 26, 2025 by ffrujeri Loading…
1 of 4 tasks
10
refactor: refactor dataset module CI:L1 Run doctests, unit tests, and functional tests documentation Improvements or additions to documentation
#977 opened Aug 25, 2025 by yuki-97 Loading…
1
Ko3n1g/tk/prebuilt wheels CI:L2 Run doctests, unit tests, functional tests, and convergence tests
#972 opened Aug 24, 2025 by ko3n1g Loading…
4 tasks
1
feat: FP8 Training in Megatron Path documentation Improvements or additions to documentation
#971 opened Aug 24, 2025 by guyueh1 Loading…
1 of 4 tasks
1
1
feat: Add NRL_CONVERT_MEGATRON_ON_ALL_LOCAL_RANK_0 option
#967 opened Aug 22, 2025 by yfw Loading…
4 tasks
5
docs: update v0.4 features + add quick start section
#965 opened Aug 22, 2025 by terrykong Loading…
4 tasks
fix: address double bos in eval task
#962 opened Aug 21, 2025 by ZhiyuLi-Nvidia Loading…
2 of 4 tasks
3
docs: guide for sliding puzzle example documentation Improvements or additions to documentation
#961 opened Aug 21, 2025 by slikhite-1 Loading…
feat: GRPO example for Qwen3 32b context length=128k CI:L1 Run doctests, unit tests, and functional tests
#957 opened Aug 20, 2025 by soodoshll Loading…
4 tasks
@soodoshll
7
fix: fix scheduler decay steps with megatron backend
#939 opened Aug 19, 2025 by ashors1 Loading…
4 tasks
1
feat: support swanlab logger CI:docs Run doctest documentation Improvements or additions to documentation
#923 opened Aug 14, 2025 by terrykong Loading…
6
feat: multi-turn search R1 example
#914 opened Aug 13, 2025 by soodoshll Loading…
4 tasks
@soodoshll
20
ci: Set submodule check to use pull_request_target
#913 opened Aug 13, 2025 by chtruong814 Loading…
4 tasks
feat: add tp_comm_bootstrap_backend,tp_comm_overlap as mcore backend …
#912 opened Aug 13, 2025 by terrykong • Draft
4 tasks
feat: Migration from NeMo Tron to Megatron Bridge
#905 opened Aug 12, 2025 by yaoyu-33 • Queued
1
37
feat: Enable global post process and metrics
#899 opened Aug 12, 2025 by jubick1337 Loading…
4 tasks
ProTip! Mix and match filters to narrow down what you’re looking for.