NVIDIA-NeMo/Nemotron

View on GitHub

Developer Asset Hub for NVIDIA Nemotron — A one-stop resource for training recipes, usage cookbooks, datasets, and full end-to-end reference examples to build with Nemotron models

Stars: 2.1k
Forks: 420
Open beginner issues: 0
Indexed issues: 37
Avg merge: 2d 15h
Merged PRs (30d): 5

Dominant language: Jupyter Notebook
License: Apache-2.0
Last GitHub push: Sep 11, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

37 open issues indexed

[Architecture Proposal] Hardware-Gated Containment Framework (Genesis Protocol V5.0) for Recursive Self-Improvement Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

NVIDIA-NeMo/Nemotron#349 · Opened Sep 13, 2026
Add optional NeMo Gym evaluation to `lightning35 pipe` Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#348 · Opened Sep 12, 2026
Lightning35 and Nano3 Slurm launchers ignore recipe resource defaults Open

Difficulty 3/5 1-2 days Newbie friendliness 78/100

NVIDIA-NeMo/Nemotron#345 · Opened Sep 10, 2026
Implement staging for rerank and embed prep/finetune; prevent embed `--stage` from following normal execution Open

Difficulty 4/5 3-5 days Newbie friendliness 72/100

NVIDIA-NeMo/Nemotron#344 · 2 comments · Opened Sep 10, 2026
[Bug] Nemotron Answering as Qwen Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#337 · Opened Aug 29, 2026
[Nemotron Embed Fine-Tuning] Log Stage 0 LLM responses and validate empty QA output Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 52/100

NVIDIA-NeMo/Nemotron#314 · Opened Aug 4, 2026
Update FastText language filtering for Curator recipe Open

Difficulty 3/5 1-2 days Newbie friendliness 55/100

NVIDIA-NeMo/Nemotron#313 · 2 comments · Opened Aug 4, 2026
Verify evals on Papers with Code Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#302 · Opened Jul 22, 2026
[Bug] [Question] Is the enable_thinking gate in boundary detection intentional for loss mask behavior? Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 45/100

NVIDIA-NeMo/Nemotron#290 · 1 comment · Opened Jul 8, 2026
[Bug] Nemotron 3 Ultra-550B-A55B-NVFP4 on 8xH100: error in cookbook + unsupported feature on sm_90a Open

bug

NVIDIA-NeMo/Nemotron#289 · 1 comment · 1 assignee · Opened Jul 7, 2026
tests: tests/recipes/omni3/test_cli.py imports missing nemotron.cli.commands.omni3.build module Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#284 · Opened Jul 1, 2026
tests: data_prep tests fail to collect due to missing cosmos_xenna dependency Open

Difficulty 3/5 1-2 days Newbie friendliness 50/100

NVIDIA-NeMo/Nemotron#281 · 2 comments · Opened Jul 1, 2026
[Ultra3 docs] MOPD/RL recipe references configs & scripts absent from all public repos Open

Difficulty 4/5 3-5 days Newbie friendliness 42/100

NVIDIA-NeMo/Nemotron#265 · 1 reaction · Opened Jun 23, 2026
Questions about datasets which are not released but used for model training Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#263 · 1 reaction · Opened Jun 23, 2026
Nemotron Super RL is using Nano Parser Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 48/100

NVIDIA-NeMo/Nemotron#259 · Opened Jun 10, 2026
Are Phase 1 weights reliable? Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#248 · Opened Jun 3, 2026
Tests and pre-commit hooks are broken Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 38/100

NVIDIA-NeMo/Nemotron#213 · Opened May 22, 2026
Request for Nemotron 3 Intermediate Checkpoints for Safety Research Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#212 · Opened May 21, 2026
Request: Intermediate training checkpoints for Nemotron 3 Nano Omni Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#200 · Opened May 11, 2026
Release Nemotron 3 Super artifacts (models, dataset) on Hugging Face Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#148 · 2 comments · 1 reaction · Opened Apr 15, 2026
nemotron-cc-code in super3 pretraining data recipe should be publicly available Open

bug

Difficulty 2/5 1-3 hours Newbie friendliness 65/100

NVIDIA-NeMo/Nemotron#146 · Opened Apr 15, 2026
Bug: beam search crashes with `HybridMambaAttentionDynamicCache` (4 bugs in `modeling_nemotron_h.py`) (NemotronH 30B A3B) Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

NVIDIA-NeMo/Nemotron#142 · Opened Apr 12, 2026
stage 2 training with CP>1 Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#137 · Opened Apr 9, 2026
Nemotron 3 Super DGX Spark Deployment Guide broken Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 48/100

NVIDIA-NeMo/Nemotron#132 · 6 comments · Opened Apr 7, 2026
Trying to replicate super v3 with large contexts Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 42/100

NVIDIA-NeMo/Nemotron#123 · 2 comments · Opened Mar 21, 2026
Datasets release plan required Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 30/100

NVIDIA-NeMo/Nemotron#120 · Opened Mar 18, 2026
Blend for Super 3 and Training Questions Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 20/100

NVIDIA-NeMo/Nemotron#119 · Opened Mar 16, 2026
How long did pretraining take? Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#114 · Opened Mar 15, 2026
Add long-context document analysis use-case example for Nemotron 3 Super Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

NVIDIA-NeMo/Nemotron#108 · 1 comment · Opened Mar 12, 2026
Add high-throughput batch processing cookbook for Nemotron 3 Super Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

NVIDIA-NeMo/Nemotron#107 · 1 comment · Opened Mar 12, 2026
Add agentic tool-calling use-case example for Nemotron 3 Super Open

Difficulty 3/5 1-2 days Newbie friendliness 55/100

NVIDIA-NeMo/Nemotron#105 · 1 comment · Opened Mar 12, 2026
Question about how Dataset.weight controls data mixing ratio in SFT preprocessing Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#78 · 1 comment · 1 reaction · Opened Mar 4, 2026
Support customized training configurations Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 15/100

NVIDIA-NeMo/Nemotron#60 · 1 comment · Opened Jan 20, 2026
Support advanced data mixing strategies Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#59 · 1 comment · Opened Jan 20, 2026
[Feature Request] Adaptive-K Routing: Dynamic Expert Selection Based on Router Entropy Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#58 · Opened Jan 19, 2026
偶发输出陷入死循环，不停重复输出 Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 25/100

NVIDIA-NeMo/Nemotron#34 · Opened Dec 18, 2025
“你是谁？“ 回答错误 Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA-NeMo/Nemotron#33 · 1 comment · 2 reactions · Opened Dec 18, 2025

NVIDIA-NeMo/Nemotron

Get new issues in your inbox