GPU-OPTIMIZED SOFTWARE HUB FOR KI, MACHINE LEARNING
AND HIGH-PERFORMANCE COMPUTING
AND HIGH-PERFORMANCE COMPUTING
eNTERPRISE CLOUD SERVICES
Features and benefits
Faster solution development
The NGC catalog increases productivity with easy-to-implement, optimized AI frameworks and HPC application containers - allowing users to focus on developing their solutions.
Simple AI introduction
The NGC catalog lowers the barrier to AI adoption, does the rough work (expertise, time, and compute resources) with pre-trained models and workflows at the highest levels of precision and performance.
Run software anywhere with
NVIDIA GPUs
Run software from the NGC catalog locally, in the cloud, peripherally, or in hybrid and multi-cloud deployments. Software from the NGC catalog can be deployed on bare metal servers or in virtualized environments, maximizing GPU utilization
and application portability and scalability.
Deploy NGC software with confidence
Run software from the NGC catalog Get direct access to NVIDIA experts with enterprise-class support for NVIDIA certified systems. This minimizes system downtime and maximizes system utilization and productivity.
PLATFORM FOR ALL APPLICATIONS
Language Modeling
Language modeling is natural language processing (NLP) that determines the probability of a particular sequence of words in a sentence.
View Language Modeling Collection
Recommendation services
Recommendation services are information filtering systems that attempt to predict the "rating" or "preference" that a user would assign to an item.
Show collection on recommendation services
Image segmentation
Image segmentation is a field of image processing in which an image is divided into several subgroups or areas representing characteristic objects or subdivisions.
View image segmentation collection
Translation
Machine translation is the automated translation of text from one language to another.
View translation collection
Object detection
Object detection is not only about recognizing the presence and position of objects in images and videos, but also categorizing them into everyday objects.
SHOW OBJECT DETECTION COLLECTION
Automatic speech recognition
Use cases for Automatic Speech Recognition (ASR) systems include issuing voice commands to interactive virtual assistants, converting audio to subtitles for online videos, and transcribing customer interactions into text for archiving in a call center.
SHOW COLLECTION FOR AUTOMATIC SPEECH RECOGNITION
Text-to-Speech
Speech synthesis, or text-to-speech transcription, means artificially generating human speech from simple transcripts. Text-to-speech models can be used, for example, when an interactive virtual assistant responds to a request or when a mobile device converts text on a web page to speech for accessibility reasons.
SHOW COLLECTION FOR SPEECH SYNTHESIS
HPC
High-performance computing (HPC) is one of the most important tools for progress in computer science. The universe of scientific computing has expanded in all directions.
SHOW COLLECTION ON HPC
READY FOR BUSINESS
Power
From deep-learning containers that are updated monthly to make the most of your GPUs' power, to cutting-edge AI models that set benchmark records in MLPerf, the NGC catalog is a critical component to achieving faster solutions and reducing time to market.
Security
Containers undergo rigorous security scans for common vulnerabilities (CVEs), crypto keys, private keys, and metadata before being published to the catalog.
More information
Privacy
NGC Private Registry provides a secure, cloud-native space where you can store and share your own containers, models, model scripts, and Helm Charts with your organization. Access to NGC Private Registry is available to customers who have purchased enterprise support with their NVIDIA DGX™ or other NVIDIA Certified Systems™.
More information
Upgrade your production with AI
that enables organizations to solve new challenges while increasing operational efficiency.
SAFE EXECUTION AT ANY LOCATION
Local
Software from the NGC catalog runs on a variety of different GPU-accelerated platforms from NVIDIA, including NVIDIA certified systems, NVIDIA DGX systems workstations running NVIDIA RTX™, and virtualized environments running NVIDIA Virtual Compute Server.
In the cloud
Software from the NGC catalog can be deployed on GPU-powered instances. The software can be deployed directly on virtual machines (VMs) or on Kubernetes services from major Cloud service providers (CSPs) deployed. With AI software from NVIDIA, enterprises can easily develop and deploy their solutions in the cloud.
In the Edge
As computing expands beyond data centers and the edge, software from the NGC catalog can be used for low-latency, high-throughput inference on Edge systems based on Kubernetes be deployed.
FASTER CREATION OF KI APPLICATIONS WITH NVIDIA TAO
NVIDIA TAO is a platform for training, customization, and optimization of AI models that eliminates the need for large training sets and extensive AI expertise and simplifies the creation of enterprise AI applications.
MORE INFORMATION
Speed up your workflow with the NGC catalog
Faster deployment and execution
of workloads with containers
The NGC Catalog hosts containers for leading AI and data science software customized, tested, and optimized by NVIDIA, as well as fully tested containers for HPC applications and data analytics. The NGC Catalog containers provide powerful and easy-to-deploy software that delivers the fastest results and allows users to develop solutions from a tested framework, with complete control.
Quick start of projects with pre-trained models
Many AI applications have the same requirements: Classification, object recognition, language translation, text-to-speech conversion, recommendation services, sentiment analysis, etc. Applications with these functions can be developed faster if you start with a pre-trained model that you optimize for your specific use case.
The NGC catalog provides pre-trained models for a variety of general AI tasks that are optimized for NVIDIA Tensor Core GPUs and can be easily re-trained by updating a few layers, saving you valuable time.
Automated deployments with Helm Charts
Helm charts automate software deployment to Kubernetes clusters and allow users to focus on using, not installing, your software.
The NGC Catalog hosts Kubernetes-ready Helm charts that simplify the deployment of powerful third-party software. The NGC Catalog also enables DevOps to push and share their Helm charts so teams can leverage consistent, secure and reliable environments to accelerate development and production cycles.
With Helm charts, the NGC Catalog offers NVIDIA GPU Operator, a suite of NVIDIA drivers, container runtime, device plug-in for the device, and management software that IT teams can install in Kubernetes clusters to give users faster access to your workloads.
DISCOVER AND IMPLEMENT ALL THE SOFTWARE YOU NEED TO BUILD KI SOLUTIONS FASTER
Collections are use case-based curated content in an easy-to-use package. Collections make it easy to discover the compatible framework containers, models, Juptyer notebooks, and other resources to get started faster. In addition, the respective Collections provide detailed documentation to deploy all content for specific use cases.
The NGC catalog provides ready-to-use Collections for various applications such as NLP, ASR, intelligent video analysis, and object recognition.
Faster deployment of solutions
with out-of-the-box AI workflows
The NGC catalog features the NVIDIA Transfer Learning Toolkit, an SDK that enables deep learning application developers and data scientists to retrain object recognition and image classification models. It can also be easily deployed with the NVIDIA DeepStream SDK for intelligent video analytics.
GETTING THE BEST OUT OF YOUR SYSTEMS
CERTIFIED BY NVIDIA
