Skip to content

Pull requests: NVIDIA-NeMo/Megatron-Bridge

New pull request New
158 Open 2,763 Closed
158 Open 2,763 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

[Qwen3.5 VL] Fix vlm_step padding for BSHD case ready-to-merge PR is approved, current, and only waiting for CI to pass before merge
#3611 opened Apr 30, 2026 by zhongbozhu Contributor Loading…
5 tasks
2
[examples] fix: use qwen3_vl_step in Qwen3.5 VL slurm scripts docs-only With great power comes great responsibility.
#3610 opened Apr 30, 2026 by cuichenx Contributor Loading…
3 tasks
add mistral common to deps and lower bound transformers to 5.5 full-test-suite
#3609 opened Apr 30, 2026 by suiyoubi Contributor Loading…
5 tasks
4
cp: fix(docker): resolve lightning not found on PyPI by providing local stub (3604) into r0.4.0 cherry-pick Run CICD
#3606 opened Apr 30, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
cp: num_layers_fix- qwen vl 235b_a22b on B200 (3589) into r0.4.0 cherry-pick Run CICD
#3603 opened Apr 30, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
hep permute fusion
#3601 opened Apr 30, 2026 by malay-nagda Contributor • Draft
5 tasks
1
cp: add VP for LoRA Lm3 70B (3547) into r0.4.0 cherry-pick Run CICD
#3596 opened Apr 30, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
chore(beep boop 🤖): Bump uv.lock (main, mcore-dev) (2026-04-30) full-test-suite
#3593 opened Apr 30, 2026 by svcnvidia-nemo-ci Contributor Loading…
3
chore(beep boop 🤖): Bump uv.lock (r0.4.0, mcore-core_r0.17.0) (2026-04-30) full-test-suite
#3591 opened Apr 30, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
fp4_param_gather=false
#3590 opened Apr 30, 2026 by malay-nagda Contributor • Draft
5 tasks
1
[peft] feat: validate PEFT target modules with deferred init waiting-on-customer Waiting on the original author to respond
#3587 opened Apr 29, 2026 by cuichenx Contributor Loading…
3 tasks
4
[training] chore: pass is_hybrid_cp=False to get_batch_on_this_cp_rank (mcore #4103)
#3584 opened Apr 29, 2026 by cuichenx Contributor • Draft
4 tasks
1
[model] feat: support step3vl area:model Model implementations and HF bridge logic feature New capabilities, enhancements, or enablement work
#3574 opened Apr 29, 2026 by shifangx Contributor Loading…
5 tasks
1
Reverted part of the code changes in pr2334 area:model Model implementations and HF bridge logic bug Something isn't working model-qwen 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
#3571 opened Apr 29, 2026 by shifangx Contributor Loading…
5 tasks
3
chore(beep boop 🤖): Bump uv.lock (r0.4.0, mcore-core_r0.17.0) (2026-04-29) full-test-suite
#3568 opened Apr 29, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
dsv4 import
#3562 opened Apr 28, 2026 by weijiac0619 Contributor • Draft
5 tasks
1
chore(beep boop 🤖): Bump uv.lock (r0.4.0, mcore-core_r0.17.0) (2026-04-28) full-test-suite
#3548 opened Apr 28, 2026 by svcnvidia-nemo-ci Contributor Loading…
2
Perf Config for 1 node GB200 DSV3 area:perf Performance optimizations and benchmarking area:recipe Training recipes and launch configs feature New capabilities, enhancements, or enablement work
#3545 opened Apr 27, 2026 by gautham-kollu Contributor Loading…
5 tasks
18
feat: add deterministic training support
#3543 opened Apr 27, 2026 by ZhiyuLi-Nvidia Contributor • Draft
5 tasks
1
Add HybridModel support to DeepSeek-V3
#3537 opened Apr 27, 2026 by janEbert Contributor • Draft
1
Update Llama3 70B LoRA GB200 BF16 perf config area:peft Parameter-efficient fine-tuning (LoRA, adapters) area:perf Performance optimizations and benchmarking
#3530 opened Apr 26, 2026 by rhmukundan Contributor • Draft
@rhmukundan
1
perf(fix): accumulate per-microbatch FLOPS metadata for accurate… area:perf Performance optimizations and benchmarking bug Something isn't working community-request waiting-on-maintainers Waiting on maintainers to respond
#3529 opened Apr 26, 2026 by SophusDavid Loading…
2 of 5 tasks
1
10
[model] feat: Add stepfun-ai/Step-3.5-Flash bridge area:model Model implementations and HF bridge logic feature New capabilities, enhancements, or enablement work help wanted Extra attention is needed
#3525 opened Apr 26, 2026 by shifangx Contributor Loading…
5 tasks
28
feat(mimo): add Whisper audio encoder to LLaVA MIMO training pipeline area:model Model implementations and HF bridge logic feature New capabilities, enhancements, or enablement work high-complexity Harder to merge: prone to conflicts and needs additional test coverage
#3520 opened Apr 24, 2026 by kamran-nvidia Contributor Loading…
5 tasks
18
[ci, recipe] test: add L1 functional coverage for qwen3_vl forward step area:recipe Training recipes and launch configs ci CI, automation, test queue, or workflow infrastructure work model-qwen needs-author Author action is required before review or merge can continue 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
#3502 opened Apr 23, 2026 by cuichenx Contributor Loading…
1 task
5
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.