代码库

NVIDIA Object Oriented Agents: the Pythonic way to build AI Agents.
Python
A scalable, agentic-first, and HuggingFace-native RL framework for research (9k lines).
Python
agentic-rlcontext-parallelismexpert-parallelismfsdphuggingfacellmrl-frameworkrlhfrlvrtensor-parallelism
Evaluate and improve models and agents using environments
Python
agentsbenchmarksenvironmentsevaluationgymllmreinforcement-learningreinforcement-learning-environmentsrl-environmentrl-training
A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)
Python
asrdeeplearninggenerative-aimachine-translationneural-networksspeaker-diariazationspeaker-recognitionspeech-synthesisspeech-translationtts
Make the agents you ship faster, more accurate, and safer.
Python
🛡️ NeMo Safe Synthesizer: Create private, safe versions of sensitive tabular datasets.
Python
🕵️ NeMo Anonymizer: Detect and protect PII through context-aware replacement and rewriting
Python
anonymization-pipelinecontextual-privacyllmnvidianvidia-nemopii-anonymizationsdg
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
Jupyter Notebook
aifine-tuningmodel-trainingnemotronnvidiareinforcement-learning
A project to improve skills of large language models
Python
Scalable toolkit for efficient model reinforcement
Python
Scalable data pre processing and curation toolkit for LLMs
Python
datadata-curationdata-prepdata-preparationdata-processingdata-processing-pipelinesdata-qualitydatacurationdatarecipesdeduplicationfast-data-processingfine-tuninglarge-language-modelslarge-scale-data-processingllmllm-data-qualityllmappspythonsemantic-deduplication
Training library for Megatron-based models with bidirectional Hugging Face conversion capability
Python
🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.
Python
agentic-aidata-augmentationdata-generationllmmcpmultimodalnemonvidiasdgsynthetic-datatool-use
🚀 Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support
Python
agentdeepseek-v4deepseek-v4-flashfinetuninggemma3gemma4glmgpt-osskimi-k3llamallama3llmminimax-m2mistralopenaiqwen-38-27bqwen3qwen3-nextvlm