Skip to content

Pull requests: NVIDIA/NeMo

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

Fix deepseek export dtype
#14307 opened Jul 22, 2025 by cuichenx Loading…
8 tasks
fix: import more specific to avoid circular dependency. core Changes to NeMo Core
#14306 opened Jul 22, 2025 by PeiyuanQi Loading…
updated with some things from NeMo main (double_buffer)
#14305 opened Jul 22, 2025 by salberdi-nvidia Loading…
8 tasks
feat: Add MFT (Minifinetuning) loss support for knowledge distillation
#14298 opened Jul 21, 2025 by pbelcak Loading…
4 of 8 tasks
Llama4 Export: Remove outdated MLP weight transform
#14297 opened Jul 21, 2025 by suiyoubi Loading…
8 tasks
Llama4 unified checkpoint export
#14296 opened Jul 21, 2025 by yueshen2016 Loading…
8 tasks
[lhotse] sharegpt data and testloader common Run CICD
#14294 opened Jul 21, 2025 by huckiyang Loading…
4 of 8 tasks
Disable CUDA Graphs for DeepSeek-V3
#14276 opened Jul 18, 2025 by scsudhakaran Draft
Rohit/pretrain grounded vlm
#14274 opened Jul 18, 2025 by rohitrango Draft
8 tasks
ProTip! Adding no:label will show everything without a label.