ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIAKit SDK - Windows (PB25h1)NVIDIAKit SDK - Linux (PB25h1)NVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator

NGC Catalog

All you need to build AI—GPU-optimized containers, pretrained models, SDKs, and Helm charts—unified in one catalog for cloud, data-center, or edge.

Top Technology

The most-downloaded AI frameworks, models, and SDKs on NGC—curated by NVIDIA, battle-tested by the community, ready to accelerate your next project.
View More
NVIDIA
NVIDIA
PyTorch
PyTorch is a GPU accelerated tensor computational framework. Functionality can be extended with common Python libraries such as NumPy and SciPy. Automatic differentiation is done with a tape-based system at the functional and neural network layer levels.
Container
NVIDIA
NVIDIA
Triton Inference Server
Triton Inference Server is an open source software that lets teams deploy trained AI models from any framework, from local or cloud storage and on any GPU- or CPU-based infrastructure in the cloud, data center, or embedded devices.
Container
NVIDIA
NVIDIA
cuda
Container registry for CUDA images
Container
NVIDIA Developer Program
NVIDIA
NVIDIA
DeepSeek-R1
NVIDIA NIM for GPU accelerated DeepSeek-R1 inference through OpenAI compatible APIs
Container
Collection
Cosmos Transfer1 7B
NVIDIA
NVIDIA
Cosmos World Foundation Models
Cosmos World Foundation Models: A family of highly performant pre-trained world foundation models purpose-built for generating physics-aware videos and world states for physical AI development.
815
Collection
NVIDIA
NVIDIA
DeepStream SDK
NVIDIA DeepStream SDK enables developers to build accelerated pipelines for a wide range of use-cases such as IVA, retail, industrial inspection, and many more with minimal development effort.
2273

NVIDIA NIM

A set of easy-to-use microservices for accelerating the deployment of foundation models on any cloud or data center and helps keep your data secure.
Documentation
View More
NVIDIA Developer Program
Meta
Meta
Llama 4 Maverick 17B 128E Instruct
This container houses Llama 4 Maverick model which is a general purpose multimodal, multilingual 128 MoE model with 17B parameters.
Container
NVIDIA Developer Program
NVIDIA
NVIDIA
mixtral-8x7b-instruct-v0-1
This container houses the Mixtral-8x7B-Instruct-v0.1, a sophisticated, instruction-tuned model from Mistral AI. It uses a Sparse Mixture of Experts (SMoE) architecture, dynamically selecting 2 of its 8 experts per token for greater efficiency.
Container
NVIDIA Developer Program
NVIDIA
NVIDIA
Qwen3-32B
Qwen3-32B is a large language model with 32 billion parameters of the Qwen3 series that delivers groundbreaking advancements in reasoning
Container
NVIDIA Developer Program
ipd
ProteinMPNN
Predicts amino acid sequences from 3D structure of proteins.
Container
NVIDIA Developer Program
NVIDIA
NVIDIA
Llama-3-Taiwan-70B-Instruct
NVIDIA NIM for GPU accelerated Llama-3-Taiwan-70B-Instruct inference through OpenAI compatible APIs
Container
NVIDIA Developer Program
NVIDIA
NVIDIA
gpt-oss-20b
NVIDIA NIM for GPU accelerated GPT-OSS-20B Model inference through OpenAI compatible APIs
Container

Try NVIDIA AI Enterprise

NVIDIA software that is portable across the cloud, data center, and edge. The software is designed to deliver optimized performance, robust security, and stability for development and production AI use cases.
Documentation
View More
NeMo Retriever
—
NVIDIA AI Enterprise
NeMo Retriever
Nemo Retriever is a collection of NIMs reference architectures for RAG
424
NVIDIA AI Enterprise Infra 7
NVIDIA
NVIDIA
NVIDIA AI Enterprise
NVIDIA AI Enterprise Infra 7
Access the latest infrastructure software and tools of the NVIDIA AI Enterprise 7 release, exclusively available with your NVIDIA AI Enterprise subscription.  This collection includes Government Ready images for regulated environments.
333
Long-Term Support Branch 2 (LTSB 2)
NVIDIA
NVIDIA
NVIDIA AI Enterprise
Long-Term Support Branch 2 (LTSB 2)
Access the Long-Term Support Branches (LTSB) of AI frameworks and SDKs. Supported for 36-months, with quarterly patches for high and critical software vulnerabilities.
51
Production Branch - May 2025 (PB 25h1)
NVIDIA
NVIDIA
NVIDIA NeMo Microservices
Production Branch - May 2025 (PB 25h1)
Access the production branches of AI frameworks and SDKs. Supported for 9 months with monthly security patches.
111

Try NVIDIA Omniverse

A collection of libraries and microservices for developing physical AI applications, such as industrial digital twins and robotics simulation.
Documentation
View More
Omniverse (PB 25h1)
NVIDIA
NVIDIA
NVIDIA Omniverse
Omniverse (PB 25h1)
Access the production branches of Omniverse frameworks and SDKs. Supported for 9 months with monthly security patches.
6716
Omniverse Kit (FB)
NVIDIA
NVIDIA
NVIDIA AI Enterprise
Omniverse Kit (FB)
Feature Branches provide cutting-edge updates, incorporating the latest Kit Application Framework, Kit Extensions, and Omniverse SDKs. These branches are regulary updated and tailored for developers needing early access to innovations.
110
Omniverse Kit App Streaming
NVIDIA
NVIDIA
NVIDIA AI Enterprise
Omniverse Kit App Streaming
Omniverse Kit App Streaming
551
USD Search API
NVIDIA
NVIDIA
NVIDIA AI Enterprise
USD Search API
AI-powered search for OpenUSD data, 3D models, images, and assets using text or image-based inputs.
12

Top Use Cases

Explore mission-ready playbooks—synthetic data generation, AI assistants, virtual factories, and more—powered by the catalog's most-downloaded models, containers, and resources.
Natural Language Processing
Natural Language Processing
Object Detection
Object Detection
Video Analytics
Video Analytics
Automatic Speech Recognition
Automatic Speech Recognition
Image Segmentation
Image Segmentation
Language Modeling
Language Modeling
Simulation and Modeling
Simulation and Modeling
Text to Speech
Text to Speech