Skip to content

Pull requests: NVIDIA-NeMo/Megatron-Bridge

New pull request New
144 Open 4,514 Closed
144 Open 4,514 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

fix(recipe): recompute Gemma 4 VL SFT layers area:recipe Training recipes and launch configs bug Something isn't working label-card Model verification card needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5517 opened Aug 11, 2026 by yaoyu-33 Contributor Loading…
5 tasks done
3
cp: docs(dsv4): preserve container packages during MCore sync (5514) into r0.6.0 cherry-pick Run CICD
#5516 opened Aug 11, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
cp: fix(recipe): prevent Qwen3.5-VL FSDP SFT OOM (5508) into r0.6.0 cherry-pick Run CICD
#5515 opened Aug 11, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
fix(model): AUT-1462 defer Kimi K3 optional runtime imports
#5512 opened Aug 11, 2026 by svcnemo-autobot Contributor Loading…
2
cp: fix(deepseek-v4): use full-prefix inference (5509) into r0.6.0 area:model Model implementations and HF bridge logic bug Something isn't working cherry-pick docs-only With great power comes great responsibility. needs-review PR is ready for code review and waiting on a reviewer r0.6.0 Auto-cherrypick to release branch. Apply before merge; cherrypick happens after merge. Run CICD
#5511 opened Aug 11, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
GPT-OSS 120B Verification Card (#5397)
#5510 opened Aug 11, 2026 by maanug-nv Contributor • Draft
5 tasks
1
fix(recipe): recompute Gemma 4 VL SFT layers area:recipe Training recipes and launch configs bug Something isn't working cherry-pick label-card Model verification card needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5506 opened Aug 11, 2026 by yaoyu-33 Contributor Loading…
5 tasks done
4
[fix] misc: properly pass global rank area:misc Cross-cutting utilities, logging, helpers, and other changes bug Something isn't working community-request needs-more-tests Requires additional L0 and L1 test coverage before merge ready-to-merge PR is approved, current, and only waiting for CI to pass before merge
#5505 opened Aug 11, 2026 by pavelgein Contributor Loading…
2 of 5 tasks
1
fix(model): handle GEMMA4_CONVERSION_MODE=text for gemma4 MoE checkpoints area:model Model implementations and HF bridge logic bug Something isn't working community-request needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer ready-to-merge PR is approved, current, and only waiting for CI to pass before merge
#5503 opened Aug 11, 2026 by Josephasafg Loading…
2
[misc] fix: Support prompt logprobs in OpenAI server area:misc Cross-cutting utilities, logging, helpers, and other changes bug Something isn't working needs-review PR is ready for code review and waiting on a reviewer
#5501 opened Aug 11, 2026 by yaoyu-33 Contributor Loading…
6
cp: chore(docker): remove unnecessary image packages area:build Dependencies, packaging, images, and environment setup cherry-pick ci CI, automation, test queue, or workflow infrastructure work full-test-suite needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5498 opened Aug 11, 2026 by svcnemo-autobot Contributor Loading…
2
[ckpt] fix: restore frozenset Energon group keys area:ckpt Checkpoint conversion, loading, export, and save paths bug Something isn't working needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5497 opened Aug 11, 2026 by yaoyu-33 Contributor Loading…
5
feat(data): add in-batch packing for GPT SFT area:data Dataset builders, preprocessing, and samplers feature New capabilities, enhancements, or enablement work needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5493 opened Aug 11, 2026 by yaoyu-33 Contributor Loading…
4 tasks done
4
fix(container): support non-root repository tests
#5486 opened Aug 10, 2026 by ko3n1g Contributor • Draft
1
[depends on mcore 6252] feat(model): support configurable logit dtype area:model Model implementations and HF bridge logic blocked Work cannot move forward until an external dependency is cleared feature New capabilities, enhancements, or enablement work
#5483 opened Aug 10, 2026 by cuichenx Contributor • Draft
5 tasks done
1
feat(model): add Muse Glimmer support
#5479 opened Aug 10, 2026 by yaoyu-33 Contributor • Draft
4 tasks done
3
fix(data): exclude packed padding from MoE router statistics area:data Dataset builders, preprocessing, and samplers area:training Training loop, callbacks, and runtime integration bug Something isn't working needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5470 opened Aug 10, 2026 by yaoyu-33 Contributor Loading…
4 tasks done
5
feat(perf): add NCCL EP GB200 Qwen3 recipes
#5468 opened Aug 10, 2026 by malay-nagda Contributor • Draft
5 tasks
1
feat(data): add Nemotron Image v3 Energon tutorial area:data Dataset builders, preprocessing, and samplers feature New capabilities, enhancements, or enablement work needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5455 opened Aug 9, 2026 by yaoyu-33 Contributor Loading…
11
feat(data): blend several sources on the direct-HF SFT path area:data Dataset builders, preprocessing, and samplers community-request feature New capabilities, enhancements, or enablement work needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5438 opened Aug 8, 2026 by bzantium Contributor Loading…
3 of 5 tasks
1
ci: AUT-1408 replace cicd PAT with GitHub App tokens area:ci ci CI, automation, test queue, or workflow infrastructure work full-test-suite needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5415 opened Aug 7, 2026 by svcnemo-autobot Contributor Loading…
8
Add missing config options for TE quantization recipes area:training Training loop, callbacks, and runtime integration feature New capabilities, enhancements, or enablement work needs-more-tests Requires additional L0 and L1 test coverage before merge needs-review PR is ready for code review and waiting on a reviewer
#5408 opened Aug 7, 2026 by timmoon10 Loading…
1 of 5 tasks
1
real routing dsv3
#5407 opened Aug 7, 2026 by malay-nagda Contributor • Draft
5 tasks
1
fix(training): preserve CLI mixed precision overrides area:training Training loop, callbacks, and runtime integration bug Something isn't working needs-review PR is ready for code review and waiting on a reviewer
#5406 opened Aug 7, 2026 by aroshanghias-nvd Contributor Loading…
1
3
fix(switch_mcore): auto-fetch pinned commit when missing locally area:build Dependencies, packaging, images, and environment setup bug Something isn't working needs-review PR is ready for code review and waiting on a reviewer
#5403 opened Aug 7, 2026 by shifangx Contributor Loading…
5 tasks
2
ProTip! Adding no:label will show everything without a label.