Skip to content

Pull requests: NVIDIA/NeMo

New pull request New
79 Open 11,012 Closed
79 Open 11,012 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: Compatibility modification of megatron-fsdp
#14593 opened Aug 27, 2025 by shjwudp Loading…
8 tasks
Support first last N layers in mxfp8
#14591 opened Aug 27, 2025 by WanZzzzzz Loading…
8 tasks
Fix kv_channels configuration for Gemma2 27b Run CICD
#14590 opened Aug 26, 2025 by ananthsub • Draft
8 tasks
Add gpt-oss lora exporter
#14589 opened Aug 26, 2025 by cuichenx • Draft
8 tasks
Fix Magpie Tests; Update default params for Decoder Context Yamls CI Run CICD TTS
#14587 opened Aug 26, 2025 by blisc Loading…
Use lhotse dataloader for ASR models to support in-manifest channel selection for multichannel recordings ASR
#14586 opened Aug 26, 2025 by racoiaws Loading…
7 tasks
1
1
Fine-tune embedding models (E5-Large-V2 and LLaMA-3.2-1B) on the allnli triplet dataset with NeMo Framework
#14584 opened Aug 26, 2025 by girihemant19 Loading…
8 tasks
[Qwen3] MoE 480B A33B model performance recipe
#14580 opened Aug 26, 2025 by gdengk • Draft
8 tasks
2
[Qwen3] Fix flops cal for dense models
#14579 opened Aug 26, 2025 by gdengk Loading…
8 tasks
Update ModelCommPGs API from megatron-core
#14578 opened Aug 25, 2025 by yaoyu-33 Loading…
8 tasks
[eval][bugfix] avoid overidden cross attention maps for multiple repeats Run CICD TTS
#14577 opened Aug 25, 2025 by XuesongYang Loading…
3
[Perf script] Llama and GPT3 perf script use mlp cast fusion
#14575 opened Aug 25, 2025 by guyueh1 Loading…
8 tasks
4
[MagpieTTS][TTS] Streaming Algorithm for MagpieTTS to 2508 TTS
#14573 opened Aug 25, 2025 by subhankar-ghosh Loading…
8 tasks
4
Making sure best_k_models=-1 doesn't delete models
#14571 opened Aug 25, 2025 by marcromeyn Loading…
8 tasks
Enabling DDP Support for LLAMA3 70B LoRa
#14570 opened Aug 25, 2025 by rhmukundan Loading…
@rhmukundan
End detection updates and Text context remapping during training Run CICD TTS
#14569 opened Aug 25, 2025 by paarthneekhara Loading…
10
Bump TE and Mcore
#14568 opened Aug 24, 2025 by chtruong814 Loading…
8 tasks
1
Adding separate SLURM jobs for Checkpoint Download, SQUAD Dataset dow…
#14564 opened Aug 23, 2025 by rhmukundan • Draft
AutoTuner in Nemo for Lepton
#14563 opened Aug 22, 2025 by prekshivyas Loading…
8 tasks
6
autoconfigurator enhancements
#14562 opened Aug 22, 2025 by prekshivyas Loading…
8 tasks
2
Add Parakeet Hybrid RNNT CTC BPE Model with Prompt support ASR
#14561 opened Aug 22, 2025 by ealbasiri Loading…
5 of 6 tasks
Randomized shard slicing for tarred data common
#14558 opened Aug 22, 2025 by pzelasko Loading…
1 of 8 tasks
15
Add example for loading
#14557 opened Aug 22, 2025 by BoxiangW • Draft
8 tasks
Stop overwriting MCore FSDP config in MegatronStrategy
#14556 opened Aug 21, 2025 by rahimftd Loading…
1 of 8 tasks
1
[uppalutkarsh/fp8] fp8 NeMo2 tutorial (applicable for both pretraining and post-training with MCore)
#14552 opened Aug 21, 2025 by uppalutkarsh Loading…
8 tasks
ProTip! Updated in the last three days: updated:>2025-08-24.