-
Notifications
You must be signed in to change notification settings - Fork 534
Pull requests: NVIDIA-NeMo/RL
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
feat(ultra): 3K NVCF + disaggregated-sandbox pipeclean recipes and launch fixes
#3876
opened Aug 27, 2026 by
hemildesai
Contributor
•
Draft
ci: AUT-1846 set CUDA family for Megatron lock generation
CI:L2
Run doctests, unit tests, functional tests, and convergence tests
CI
Relating to CI
#3870
opened Aug 27, 2026 by
svcnemo-autobot
Contributor
Loading…
feat(rollout): integrate MInf ledger capture with checkpointing
CI:L1
Run doctests, unit tests, and functional tests
Documentation
Improvements or additions to documentation
ci: Bump Megatron-Bridge to 7664d56
CI:L1
Run doctests, unit tests, and functional tests
#3868
opened Aug 27, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
perf(refit): use NCCL Reshard for all async MXFP8 recipes
CI:Lfast
Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version)
#3865
opened Aug 27, 2026 by
seonjinn
Contributor
Loading…
perf: overlap non-colocated MInf + polict init
CI:L1
Run doctests, unit tests, and functional tests
#3864
opened Aug 27, 2026 by
tdene
Contributor
Loading…
4 tasks
ci: AUT-2161 route external jobs through ephemeral v2
CI:L0
Run doctests and unit tests
CI
Relating to CI
#3863
opened Aug 27, 2026 by
svcnemo-autobot
Contributor
Loading…
fix(ci): set CUDA major for torch-memory-saver lock generation
CI:L1
Run doctests, unit tests, and functional tests
CI
Relating to CI
#3862
opened Aug 27, 2026 by
balasaajay
Loading…
4 tasks
fix(genrm): propagate prompt cohort retry identity
#3861
opened Aug 27, 2026 by
macandro96
Contributor
Loading…
4 tasks
build: mount GITHUB_TOKEN secret for apptainer dependency download
#3860
opened Aug 26, 2026 by
kajalj22
Contributor
Loading…
2 tasks
fix(megatron): pass delegation flags on the presharded train path
#3859
opened Aug 26, 2026 by
rohitrango
Contributor
Loading…
fix(vllm): realign multimodal placeholders after prefix replacement
CI:L1
Run doctests, unit tests, and functional tests
#3858
opened Aug 26, 2026 by
aroshanghias-nvd
Contributor
Loading…
4 tasks done
feat(megatron): support GTP weight rematerialization for training + refit
#3857
opened Aug 26, 2026 by
shanmugamr1992
Contributor
•
Draft
feat: forward rollout session IDs to Dynamo
CI:L0
Run doctests and unit tests
#3856
opened Aug 26, 2026 by
jthomson04
Contributor
•
Draft
3 of 4 tasks
fix(ppo): reject refit_transport on every backend, not only on vLLM
community-request
#3855
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
fix(algorithms): raise instead of asserting on the skipped-reference-KL pairing
community-request
#3854
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
fix(algorithms): honour the configured KL clamps in the reward-side KL
community-request
#3853
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
feat: super v3.5 posttraining with flexible length penalties
#3852
opened Aug 26, 2026 by
arendu
Contributor
Loading…
fix(loss): count num_valid_samples from the sample mask, not the batch dim
community-request
#3850
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
test(sc): add the distillation recipe and functional test for SingleController
community-request
#3849
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
feat(sc): run the distillation teacher in the SingleController train pump
community-request
#3846
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
fix(vllm): return HTTP 400 for chat completion context overflow
#3844
opened Aug 26, 2026 by
jepio
Contributor
Loading…
3 of 4 tasks
feat(sc): add the teacher top-k forward for distillation on SingleController
community-request
#3843
opened Aug 26, 2026 by
tianyi-zhang-02
Contributor
Loading…
3 of 4 tasks
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.