Skip to content

Pull requests: NVIDIA-NeMo/Megatron-Bridge

New pull request New
186 Open 3,859 Closed
186 Open 3,859 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

feat(training): add shared recipe launcher
#4856 opened Jul 14, 2026 by yaoyu-33 Contributor Loading…
5
feat(recipes): add standalone NeMoDiag V0 workloads
#4855 opened Jul 14, 2026 by yaoyu-33 Contributor Loading…
4 tasks done
3
feat(recipes): add Qwen3.5 text pretraining
#4854 opened Jul 13, 2026 by cuichenx Contributor • Draft
1
fix(recipe): add FC LoRA targets for Nemotron Omni
#4853 opened Jul 13, 2026 by cuichenx Contributor • Draft
4 tasks done
1
fix(training): allow local CUDA graphs for inference scopes area:perf Performance optimizations and benchmarking bug Something isn't working ready-to-merge PR is approved, current, and only waiting for CI to pass before merge
#4851 opened Jul 13, 2026 by HollowMan6 Member Loading…
2 of 5 tasks
4
add qwen3 & deepseek-v3 moe perf recipes area:perf Performance optimizations and benchmarking 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
#4850 opened Jul 13, 2026 by dimapihtar Contributor Loading…
5 tasks
10
Add mlperf llama2 70b lora performance script
#4849 opened Jul 13, 2026 by nv-mollys Contributor • Draft
5 tasks
1
[model] fix: adapt Gemma 4 MoE to MCore packed sequences area:model Model implementations and HF bridge logic bug Something isn't working 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
#4846 opened Jul 13, 2026 by yaoyu-33 Contributor Loading…
3
chore(beep boop 🤖): Bump uv.lock (r0.5.0, mcore-core_r0.18.0) (2026-07-13) full-test-suite
#4841 opened Jul 13, 2026 by svcnvidia-nemo-ci Contributor Loading…
3
chore(beep boop 🤖): Bump uv.lock (main, mcore-dev) (2026-07-13) full-test-suite
#4840 opened Jul 13, 2026 by svcnvidia-nemo-ci Contributor Loading…
3
[training, ckpt] fix: Reset MCore async queue during restart 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
#4839 opened Jul 13, 2026 by yaoyu-33 Contributor Loading…
3
Migrate Qwen3-ASR and Qwen3-Omni to HybridModel needs-more-tests Requires additional L0 and L1 test coverage before merge
#4838 opened Jul 12, 2026 by Phlip79 Member • Draft
1
Migrate Qwen VL models to HybridModel full-test-suite needs-more-tests Requires additional L0 and L1 test coverage before merge
#4837 opened Jul 12, 2026 by Phlip79 Member • Draft
1
Migrate Qwen3-Next and Qwen3.5 models to HybridModel needs-more-tests Requires additional L0 and L1 test coverage before merge
#4836 opened Jul 12, 2026 by Phlip79 Member • Draft
1
fix(data): honor --packed_sequence for llm-finetune-preloaded area:data Dataset builders, preprocessing, and samplers bug Something isn't working community-request needs-review PR is ready for code review and waiting on a reviewer
#4835 opened Jul 12, 2026 by Achyuthan-S Contributor Loading…
1
2
chore(beep boop 🤖): Bump uv.lock (main, mcore-main) (2026-07-12) full-test-suite needs-more-tests Requires additional L0 and L1 test coverage before merge
#4826 opened Jul 12, 2026 by svcnvidia-nemo-ci Contributor Loading…
6
chore(beep boop 🤖): Bump uv.lock (main, mcore-dev) (2026-07-12) full-test-suite
#4825 opened Jul 12, 2026 by svcnvidia-nemo-ci Contributor Loading…
3
docs(recipe): add DeepSeek-V4-Pro GB300 reproduction guide area:perf Performance optimizations and benchmarking docs Documentation-only updates or documentation debt docs-only With great power comes great responsibility. needs-review PR is ready for code review and waiting on a reviewer performance
#4824 opened Jul 12, 2026 by dingqingy-nv Contributor Loading…
3
chore(beep boop 🤖): Bump uv.lock (r0.5.0, mcore-core_r0.18.0) (2026-07-11) full-test-suite
#4816 opened Jul 11, 2026 by svcnvidia-nemo-ci Contributor Loading…
4
chore(beep boop 🤖): Bump uv.lock (main, mcore-main) (2026-07-11) full-test-suite
#4815 opened Jul 11, 2026 by svcnvidia-nemo-ci Contributor Loading…
3
chore(beep boop 🤖): Bump uv.lock (main, mcore-dev) (2026-07-11) full-test-suite
#4813 opened Jul 11, 2026 by svcnvidia-nemo-ci Contributor Loading…
3
test(training): validate text-only pretrain recipes area:recipe Training recipes and launch configs ci CI, automation, test queue, or workflow infrastructure work full-test-suite needs-review PR is ready for code review and waiting on a reviewer
#4810 opened Jul 11, 2026 by yaoyu-33 Contributor • Draft
5
chore: update transformer-engine dependency to version 2.17.0 area:build Dependencies, packaging, images, and environment setup 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
#4807 opened Jul 10, 2026 by balasaajay Contributor Loading…
5 tasks
10
Add mlperf llama2 70b lora performance script
#4806 opened Jul 10, 2026 by nv-mollys Contributor • Draft
5 tasks
1
[model, docs] test: validate full-model conversion and inference examples area:model Model implementations and HF bridge logic docs Documentation-only updates or documentation debt
#4805 opened Jul 10, 2026 by yaoyu-33 Contributor • Draft
67 tasks
5
ProTip! Adding no:label will show everything without a label.