Skip to content

Pull requests: NVIDIA-NeMo/Megatron-Bridge

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
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

[training] fix: guard mcore-dev-incompatible APIs
#3333 opened Apr 15, 2026 by yaoyu-33 Contributor Loading…
2 tasks done
[model] fix: Qwen3.5-VL MTP standard attn specs patch community-request
#3330 opened Apr 14, 2026 by HollowMan6 Contributor Loading…
2 of 5 tasks
[model] refactor: formalize hf_config on MegatronModelBridge full-test-suite
#3329 opened Apr 14, 2026 by yaoyu-33 Contributor Loading…
4 tasks
[model] fix: improve apply_rope_fusion assert message for Qwen3-VL docs-only With great power comes great responsibility.
#3328 opened Apr 14, 2026 by yaoyu-33 Contributor Loading…
2 tasks
Update Qwen3-VL pretrain perf configs for 30B and 235B 26.04.01 area:perf Performance optimizations and benchmarking performance/release Performance items related with NeMo release performance r0.4.0 Auto-cherrypick to release branch. Apply before merge; cherrypick happens after merge.
#3327 opened Apr 14, 2026 by tomlifu Contributor Loading…
5 tasks
ci: add sync-skills workflow
#3325 opened Apr 14, 2026 by ko3n1g Contributor Loading…
qwen_vl_2604_functional_fixes
#3322 opened Apr 14, 2026 by malay-nagda Contributor Draft
5 tasks
[model]Add Qwen3‑Omni training support community-request
#3317 opened Apr 14, 2026 by hbhflw2000 Loading…
3 of 5 tasks
[recipe] feat: add Qwen3-0.6B 128K SFT recipe with YaRN RoPE scaling
#3316 opened Apr 14, 2026 by RayenTian Contributor Loading…
5 tasks
[DSV3] Fix the ckpt loading issue when no MoE layer on the mtp rank
#3315 opened Apr 14, 2026 by gdengk Contributor Loading…
5 tasks
[docs] feat: update model support skill with encapsulation guidance docs-only With great power comes great responsibility.
#3313 opened Apr 14, 2026 by cuichenx Contributor Loading…
2 tasks
[training] fix: use int64 for TrainState counters to prevent overflow
#3312 opened Apr 14, 2026 by yaoyu-33 Contributor Loading…
1 of 2 tasks
perf(nsys): reduce CPU-side overhead in profiling defaults
#3311 opened Apr 13, 2026 by dingqingy-nv Contributor Draft
3 tasks
[ckpt] feat: support MSC for fsdp_dtensors area:ckpt Checkpoint conversion, loading, export, and save paths community-request ready-to-merge PR is approved, current, and only waiting for CI to pass before merge
#3300 opened Apr 13, 2026 by pavelgein Contributor Loading…
3 of 5 tasks
chore(beep boop 🤖): Bump uv.lock (main, mcore-dev) (2026-04-13) area:build Dependencies, packaging, images, and environment setup full-test-suite needs-review PR is ready for code review and waiting on a reviewer
#3297 opened Apr 13, 2026 by svcnvidia-nemo-ci Contributor Loading…
[model, training] fix: align Qwen3-VL padding for HybridEP community-request needs-review PR is ready for code review and waiting on a reviewer x-shopee
#3294 opened Apr 13, 2026 by neiblegy Loading…
5 tasks
ProTip! Follow long discussions with comments:>50.