Deep Learning Containers

Truly Scalable Deep Learning

NVIDIA GPU Cloud (NGC) gives you the performance and flexibility to take on your most challenging deep learning projects. It brings the power of NVIDIA AI to every industry, helping scientists and researchers to rapidly build, train, and deploy neural network models and meet evolving demands. This includes a catalog of integrated and optimized deep learning software containers that take full advantage of NVIDIA GPUs, in the cloud or on-premises.

Three Reasons Why

INNOVATION FOR EVERY INDUSTRY
Data scientists and researchers can quickly tap into the power of NVIDIA AI with the world’s fastest GPU architecture and optimized deep learning framework containers available on NGC. NVIDIA AI is helping to solve some of the most complex problems facing humankind. Providing early detection and finding cures for infectious diseases, reducing traffic fatalities, finding imperfections in critical infrastructure, delivering deep business insights from large-scale data, and much more. Every industry from automotive and healthcare to fintech is being transformed by NVIDIA AI.
TIME TO SAY GOODBYE TO DIY
These pre-integrated, GPU-accelerated containers include NVIDIA^® CUDA^® runtime, NVIDIA libraries, and an operating system. They are tuned, tested, and certified by NVIDIA to run on NVIDIA DGX^™ systems, select NVIDIA TITAN and NVIDIA Quadro GPUs, and on NVIDIA GPUs in the cloud with Alibaba, Amazon, Google, Microsoft, and Oracle.
ACCESS FROM ANYWHERE
The catalog of NVIDIA optimized deep learning software containers is available to everyone—at no cost—for use across supported platforms, from the desktop, to the data center, to the cloud. Containerized software allows for portability of deep learning jobs across environments, reducing the overhead typically required to scale AI workloads. Developers and data scientists experimenting on an NVIDIA DGX Station or a PC with the NVIDIA Pascal-powered TITAN GPU, enterprises with NVIDIA DGX-1 in the datacenter, or organizations using NVIDIA GPUs in the cloud now have access to a consistent, optimized set of tools. With NGC, researchers can spend less time on IT and more time experimenting, gaining insights, and driving results with deep learning.

Download the NGC Deep Learning Frameworks Brief

DOWNLOAD THE NGC DEEP LEARNING SOFTWARE BRIEF

Learn more about the optimizations to the top deep learning frameworks such as TensorFlow, PyTorch, MXnet, Theano, Caffe2, Microsoft Cognitive Toolkit (CNTK), and more, in this brief.

Download Now

Watch This Video to Get Started with NVIDIA GPU Cloud and AWS

Get Started with NVIDIA GPU Cloud and Amazon EC2

NVIDIA GPU Cloud makes it easy to leverage GPU-optimized deep learning frameworks on-premises or in the cloud. See how to get started quickly with NGC and Amazon Elastic Compute Cloud (Amazon EC2).

Watch Video >

FREQUENTLY ASKED QUESTIONS

What do I get for deep learning by signing up for NGC?

You get access to a comprehensive catalog of fully integrated and optimized deep learning framework containers—all at no cost.
Who can access the containers on NGC?

Anyone who signs up for an NGC account. By maintaining an account, users can download and use the latest versions of the GPU-optimized containers on supported platforms.
What is in the deep learning containers?

Each container has the NVIDIA GPU Cloud Software Stack, a pre-integrated stack of GPU-accelerated software optimized for deep learning on NVIDIA GPUs. It includes a Linux OS, CUDA runtime, required libraries, and the chosen framework or application (TensorFlow, NVCaffe, NVIDIA DIGITS, etc.)—all tuned to work together immediately with no additional setup.
What deep learning software is available on NGC?

The NGC container registry has NVIDIA GPU-accelerated releases of the most popular frameworks: NVCaffe, Caffe2, Microsoft Cognitive Toolkit (CNTK), NVIDIA DIGITS, MXNet, PyTorch, TensorFlow, Theano, Torch, CUDA (base level container for developers), as well as NVIDIA TensorRT inference accelerator.
What are the supported platforms for the deep learning containers?

The GPU-accelerated containers are tuned, tested, and certified by NVIDIA to run on NVIDIA DGX™ systems, NVIDIA TITAN (powered by NVIDIA Volta and NVIDIA Pascal™), and NVIDIA Quadro GV100, GP100 and P6000, and on NVIDIA GPUs in the cloud with Alibaba, Amazon, Google, Microsoft, and Oracle.
Can I run the deep learning containers from NGC on my PC with a GeForce® GTX 1080 Ti?

Yes, the terms of use allow the NGC deep learning containers to be used on desktop PCs running NVIDIA Volta or Pascal-powered GPUs.
How do I run deep learning containers from NGC on a cloud service provider such as Amazon or Google Cloud?

The deep learning containers on NGC are designed to run on NVIDIA Volta or Pascal™ powered GPUs. The cloud service providers (CSP) supported by NGC offer instance types that provide the appropriate NVIDIA GPUs for running the NGC containers. In order to run the containers, you will need to choose one of these instance types, instantiate the appropriate image file on it, and then access NGC from within that image. The steps to do this varies by CSP, but you can find step-by-step instructions for each CSP in the NVIDIA GPU Cloud documentation.
How often are the deep learning framework containers updated?

Monthly. The deep learning containers on NGC benefit from continuous R&D investment by NVIDIA and joint engineering with framework engineers to ensure that each deep learning framework is tuned for the fastest training possible. NVIDIA engineers continually optimize the software, delivering monthly container updates to ensure that your deep learning investment reaps greater returns over time.
What kind of support does NVIDIA offer for these deep learning containers?

Users get access to the NVIDIA DevTalk Developer Forum https://devtalk.nvidia.com, supported by a large community of deep learning and GPU experts from the NVIDIA customer, partner, and employee ecosystem.
Why is NVIDIA providing these deep learning containers?

NVIDIA is accelerating the democratization of AI by giving deep learning researchers and developers simplified access to GPU-accelerated deep learning frameworks. This makes it easy for them to run these optimized frameworks on NVIDIA GPUs in the cloud or on local systems.
If there’s no charge for the deep learning containers, do I pay for compute time?

There is no charge for the containers on the NGC container registry (subject to the terms of the TOU). However, each cloud service provider will have their own pricing for accelerated computing services.
What are the terms of use for NGC?

Please see https://ngc.nvidia.com/legal/terms

Deep Learning Containers

Truly Scalable Deep Learning

Three Reasons Why

INNOVATION FOR EVERY INDUSTRY

TIME TO SAY GOODBYE TO DIY

ACCESS FROM ANYWHERE

DOWNLOAD THE NGC DEEP LEARNING SOFTWARE BRIEF

Get Started with NVIDIA GPU Cloud and Amazon EC2

FREQUENTLY ASKED QUESTIONS

What do I get for deep learning by signing up for NGC?

Who can access the containers on NGC?

What is in the deep learning containers?

What deep learning software is available on NGC?

What are the supported platforms for the deep learning containers?

Can I run the deep learning containers from NGC on my PC with a GeForce® GTX 1080 Ti?

How do I run deep learning containers from NGC on a cloud service provider such as Amazon or Google Cloud?

How often are the deep learning framework containers updated?

What kind of support does NVIDIA offer for these deep learning containers?

Why is NVIDIA providing these deep learning containers?

If there’s no charge for the deep learning containers, do I pay for compute time?

What are the terms of use for NGC?

GET ACCESS TO GPU-ACCELERATED DEEP LEARNING CONTAINERS WITH NGC

Get the latest news on NVIDIA GPU Cloud

GET THE LATEST DATA CENTER NEWS

ORDER YOUR NVIDIA DGX STATION TODAY

The NVIDIA DGX-1 is available for purchase in select countries

The NVIDIA DGX-1 is available
for purchase in select countries