Skip to content

Pull requests: NVIDIA-NeMo/RL

New pull request New
364 Open 2,069 Closed
364 Open 2,069 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

perf(recipes): use FlashInfer for generation
#3461 opened Aug 2, 2026 by seonjinn Contributor • Draft
3 of 7 tasks
1
feat(sc): gate-authoritative token capture on the Gym tokidcap stack CI:L1 Run doctests, unit tests, and functional tests Documentation Improvements or additions to documentation
#3456 opened Aug 1, 2026 by pthombre Contributor • Draft
1
feat: FT & R
#3454 opened Jul 31, 2026 by asolergi-nv Loading…
4 tasks
2
feat: nccl v2.30.7 bump CI:L2 Run doctests, unit tests, functional tests, and convergence tests
#3453 opened Jul 31, 2026 by youngeunkwon0405 Contributor Loading…
4 tasks
@youngeunkwon0405
2
perf(modelopt): skip redundant weight fake-quant during frozen-weight logprobs community-request Documentation Improvements or additions to documentation waiting-on-maintainers Waiting on maintainers to respond
#3441 opened Jul 31, 2026 by babyplutokurt Contributor Loading…
3 of 4 tasks
1
2
fix: Make package version detection robust for release tests (closes #2816) community-request waiting-on-maintainers Waiting on maintainers to respond
#3439 opened Jul 31, 2026 by botbikamordehai2-sketch Loading…
1
1
perf: enable HybridEP in H100/B200 performance recipes CI:Lfast Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version)
#3438 opened Jul 31, 2026 by seonjinn Contributor Loading…
3
fix(vllm): don't require fastapi/uvicorn to import nemo_rl.algorithms.grpo community-request waiting-on-maintainers Waiting on maintainers to respond
#3437 opened Jul 31, 2026 by tianyi-zhang-02 Contributor Loading…
3 tasks done
1
perf: enable HybridEP on H100/B200 x86 systems CI:Lfast Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version)
#3436 opened Jul 31, 2026 by seonjinn Contributor Loading…
@seonjinn
3
fix(async): preserve rollout lookahead on resume
#3435 opened Jul 31, 2026 by yfw Contributor Loading…
4 tasks
1
chore(examples): add Super Omni multi-turn MOPD recipe with image tools
#3430 opened Jul 30, 2026 by nvilia • 3/3 Loading…
2 of 4 tasks
1
feat(sc): add checkpoint save/restore to SingleController
#3429 opened Jul 30, 2026 by haitian-nvidia Loading…
3 of 4 tasks
1
feat: add single controller doc and gb200 nightly tests CI:docs Run doctest Documentation Improvements or additions to documentation
#3427 opened Jul 30, 2026 by yuki-97 Contributor Loading…
2
feat(sglang): NVFP4 and MXFP8 quantized rollouts with Megatron weight refit community-request
#3426 opened Jul 30, 2026 by xiuhu17 Contributor • Draft
1
feat: [DONT MERGE] Fake PR for CI on opt/dev branch CI:L1 Run doctests, unit tests, and functional tests Documentation Improvements or additions to documentation
#3424 opened Jul 30, 2026 by Superjomn Contributor Loading…
4 tasks
2
chore: bump TQ to v0.19 CI:L1 Run doctests, unit tests, and functional tests
#3423 opened Jul 30, 2026 by macandro96 Contributor Loading…
4 tasks
3
[#3342][test] Chat completion parity tests for TRTLLM v.s. vLLM
#3422 opened Jul 30, 2026 by hchings Contributor • Draft
4 tasks
@hchings
1
feat(sc): gate-authoritative token capture for async Gym rollouts (token-in/token-out) CI Relating to CI Documentation Improvements or additions to documentation
#3421 opened Jul 30, 2026 by pthombre Contributor • Draft
1
fix: prevent PP hang in MoE aux-loss metric reduction community-request waiting-on-maintainers Waiting on maintainers to respond
#3419 opened Jul 29, 2026 by dafu-wu Loading…
1 of 4 tasks
2
fix(trtllm): Add TRTLLM SWE training example, update refit APIs CI:L0 Run doctests and unit tests Documentation Improvements or additions to documentation
#3418 opened Jul 29, 2026 by hchings Contributor Loading…
4 tasks
@hchings
9
feat: add NuminaMath-1.5 response dataset community-request Documentation Improvements or additions to documentation waiting-on-maintainers Waiting on maintainers to respond
#3417 opened Jul 29, 2026 by tianyi-zhang-02 Contributor Loading…
3 tasks done
1
2
Multimodal nemo gym compatible grpo pipeline
#3414 opened Jul 29, 2026 by rohitrango Contributor • 2/3 Loading…
2 of 4 tasks
2
chore: update claude review to FW-CI-templates v1.8.5 CI:docs Run doctest CI Relating to CI
#3413 opened Jul 29, 2026 by kajalj22 Contributor Loading…
1
feat(ppo): support async ppo CI:L1 Run doctests, unit tests, and functional tests Documentation Improvements or additions to documentation
#3410 opened Jul 29, 2026 by bg51717 Contributor Loading…
4 tasks done
2
feat: Qwen-Image flow-GRPO diffusion training recipe community-request Documentation Improvements or additions to documentation waiting-on-maintainers Waiting on maintainers to respond
#3408 opened Jul 29, 2026 by triple-mu Loading…
4 tasks done
1
ProTip! Follow long discussions with comments:>50.