Skip to content

Pull requests: NVIDIA-NeMo/Megatron-Bridge

New pull request New
191 Open 1,916 Closed
191 Open 1,916 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

Suppress noisy _extra_state warnings during checkpoint loading
#2689 opened Mar 7, 2026 by cuichenx Loading…
5 tasks
4
update merge_lora script
#2688 opened Mar 6, 2026 by liding-nv Loading…
6
GPT OSS fp8 recipes
#2687 opened Mar 6, 2026 by weijiac0619 Loading…
5 tasks
5
[ckpt] refactor: Consolidate fused expert mappings and fix MTP inference
#2685 opened Mar 6, 2026 by yaoyu-33 Loading…
1 of 7 tasks
7
Fix Energon data loading incompatibility with updated Qwen3-VL finetuning pipeline community-request
#2680 opened Mar 6, 2026 by aub123 Loading…
3
chore(beep boop 🤖): Bump uv.lock (main) (2026-03-06)
#2677 opened Mar 6, 2026 by svcnvidia-nemo-ci Loading…
2
chore(beep boop 🤖): Bump uv.lock (r0.3.0) (2026-03-06)
#2676 opened Mar 6, 2026 by svcnvidia-nemo-ci Loading…
2
nsys perf and eval
#2675 opened Mar 6, 2026 by malay-nagda Loading…
5 tasks
3
dsv3 no_non_det_algo
#2673 opened Mar 6, 2026 by malay-nagda Loading…
5 tasks
2
Introduce refactored GPT builder enhancement New feature or request
#2671 opened Mar 6, 2026 by maanug-nv Loading…
5 tasks
@maanug-nv
6
Unify bf16 gb300 qwen3 235b mapping performance/release Performance items related with NeMo release performance r0.3.0 Cherry-pick label for r0.3.0 release branch
#2670 opened Mar 5, 2026 by dingqingy-nv Loading…
2
Update Qwen3 235B B300 Configs to match Qwen3 B200 Configs performance/release Performance items related with NeMo release performance r0.3.0 Cherry-pick label for r0.3.0 release branch
#2669 opened Mar 5, 2026 by rhmukundan Loading…
@rhmukundan
3
Mup scaling
#2666 opened Mar 5, 2026 by mehraakash Loading…
5 tasks
6
[training] fix: validate hierarchical_context_parallel_sizes for a2a+p2p cp_comm_type
#2665 opened Mar 5, 2026 by yaoyu-33 • Draft
3 tasks
2
fix: Add get_hf_tokenizer_kwargs to NemotronHBridge community-request
#2664 opened Mar 5, 2026 by shanecmoran Loading…
1
4
chore(beep boop 🤖): Bump uv.lock (r0.3.0) (2026-03-05)
#2660 opened Mar 5, 2026 by svcnvidia-nemo-ci Loading…
2
chore(beep boop 🤖): Bump uv.lock (main) (2026-03-05)
#2659 opened Mar 5, 2026 by svcnvidia-nemo-ci Loading…
3
Override moe_token_dispatcher_type to "alltoall" when export megatron… community-request
#2658 opened Mar 5, 2026 by jaeminh Loading…
5 tasks
6
[bridge] Fix off-by-one in sliding window size for Gemma2, Gemma3, Mistral, and GPT-OSS
#2656 opened Mar 5, 2026 by cuichenx Loading…
1 of 5 tasks
3
Add Qwen 3.5 recipes
#2654 opened Mar 5, 2026 by cuichenx Loading…
4 of 5 tasks
15
feat: Add first-class Comet ML experiment tracking community-request
#2653 opened Mar 4, 2026 by shanecmoran Loading…
1
10
[misc] fix: Improve compare.py robustness for multi-GPU and vocab-padded models
#2646 opened Mar 4, 2026 by yaoyu-33 Loading…
1 of 2 tasks
9
Wan features verifications
#2645 opened Mar 4, 2026 by huvunvidia Loading…
5 tasks
1
Store hf_pretrained as properties of Megatron*Bridge classes community-request
#2644 opened Mar 4, 2026 by HollowMan6 Loading…
2 of 5 tasks
10
fix: GPTSFTChatDataset.collate_fn use pad_seq_length_to_mult instead of hardcoded 16 community-request
#2643 opened Mar 4, 2026 by shanecmoran Loading…
1
6
ProTip! What’s not been updated in a month: updated:<2026-02-06.