NVIDIA AI Enterprise
NVIDIA AI Enterprise is a suite of 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. It consists of two types of software: Application Software for building AI agents, generative AI, and many other types of AI workflows, and Infrastructure Software such as NVIDIA GPU and Networking drivers and Kubernetes Operators to help optimize and manage the use of hardware accelerators by AI, data science, and HPC applications.
Application Software
- NVIDIA NIM and NVIDIA NeMo microservices enhance model performance and speed time to deployment for generative AI.
- SDKs, frameworks, and libraries across many domains, including speech AI, route optimization, cybersecurity, and vision AI.
- Inference and training runtimes including NVIDIA Triton Inference Server, with multiple optimized backends.
- NVIDIA GPU and NVIDIA Network drivers to run in bare metal or virtualized configurations.
- Kubernetes operators for deployment and lifecycle management of NVIDIA GPU, NVIDIA Networking, NVIDIA NIM, and NVIDIA NeMo.
- Base Command Manager to streamline cluster provisioning and infrastructure monitoring.
Customers who build AI applications or operate AI Infrastructure can use whichever parts of the suite that they wish. NVIDIA has partnered with leading industry partners to ensure integration of NVIDIA software with their platform.
NVIDIA software is available as different branches to meet the needs of the individual use case.
| Software Layer | Description | Components Included | Support Matrix | Documentation |
| Infrastructure Software | The infrastructure software includes various components that ensure efficient deployment, management, and scaling of AI applications. | Infrastructure Software | Infra Support Matrix | Infra Lifecycle documentation |
| Application Software: Production Branch | A Production Branch (PB) contains production-ready AI frameworks and SDK branches to provide API stability and a secure environment for building mission-critical AI applications. | Application Production Branch (PB) Software | Production Branch (PB) documentation | Production Branch (PB) Lifecycle documentation |
| Application Software: Feature Branch | Feature Branches (FB) deliver the latest versions of NVIDIA-built and NVIDIA-optimized AI frameworks, NVIDIA NIM microservices, pre-trained models, and SDKs and provide access to the most recent software features and optimizations. | Application Feature Branch (FB) Software | Feature Branch (FB) documentation | Feature Branch (FB) Lifecycle documentation |
| Application Software: Long-Term Support Branch | A Long-Term Support Branch (LTSB) contains long-term supported AI frameworks and SDKs that provide 36 months of API stability and a secure environment for highly regulated industries. | Application Long Term Supported Branch (LTSB) Software | Long-Term Support Branch (LTSB) documentation | Long-Term Support Branch (LTSB) Lifecycle documentation |
Getting Started with NVIDIA AI Enterprise
Learn more about the NVIDIA AI Enterprise cloud-native software platform.
Get started with the NVIDIA AI Enterprise Infrastructure Software.
The goal of this document is to help organizations choose an NVIDIA AI Enterprise Release Branch, as well as provide information on how to move from one branch to another.
NVIDIA AI Enterprise Software on NGC
NVIDIA’s NGC Catalog offers a wide array of easy-to-deploy microservices, pre-trained models, and other packaged software components exclusive to NVAIE subscribers, with NVIDIA’s support guarantees.
Easy-to-use microservices designed to accelerate generative AI implementation in enterprises. Exclusively available to NVIDIA AI Enterprise Essentials subscribers, with NVIDIA's Enterprise support team's backing.
Access the latest infrastructure software and tools of the NVIDIA AI Enterprise release. The Infra Release of NVIDIA AI Enterprise offers infrastructure management and orchestration software for infrastructure and platform teams to efficiently manage and scale AI workloads.
NVIDIA AI Enterprise Licensing and Support
This guide covers the entitlement, packaging, and licensing of NVIDIA AI Enterprise. It is intended to be a quick reference to understand the product at a high level, with the corresponding SKU information.
NVIDIA Enterprise Support and Services Guide provides information for using NVIDIA Enterprise Support and services. This document is intended for NVIDIA’s potential and existing enterprise customers. This User Guide is a non-binding document and should be utilized to obtain information for NVIDIA Enterprise branded support and services.
NVIDIA AI Enterprise software products can be purchased as either perpetual licenses with support services or as annual subscriptions.
NVIDIA AI Enterprise Production Branches & Long-Term Supported Branches Release Documentation
Active Production Branches (PB) & Long-Term Supported Branches (LTSB)
| Software Branch | Compatible Infra Release | First Planned Release | Last Planned Release | Planned EOL |
| Production Branch (PB) May 2025 (PB 25h1) | NVIDIA AI Enterprise Infra Release 6.x | May 2025 | December 2025 | January 2026 |
| Production Branch (PB) October 2024 (PB 24h2) | NVIDIA AI Enterprise Infra Releases 5.x and 6.x | October 2024 | June 2025 | July 2025 |
| Long-Term Supported Branch 2 (LTSB 2) | NVIDIA AI Enterprise Infra Release 4.x | November 2024 | August 2026 or later | October 2026 or later |
All Production Branches & & Long-Term Supported Branches (LTSB)
| Software Branch | Compatible Infra Release | First Planned Release | Last Planned Release | Planned EOL |
| Production Branch (PB) May 2024 (PB 24h1) | NVIDIA AI Enterprise Infra Releases 5.x and 4.x | May 2024 | December 2024 | January 2025 |
| Production Branch (PB) October 2023 (PB 23h2) | NVIDIA AI Enterprise Infra Release 3.x | October 2023 | June 2024 | July 2024 |
| Long-Term Supported Branch 1 (LTSB 1) | NVIDIA AI Enterprise Infra Release 1.x | August 2021 | February 2024 | June 2024 |
NVIDIA AI Enterprise Application Release Branches
A Production Branch (PB) contains production-ready AI frameworks and SDK branches to provide API stability and a secure environment for building mission-critical AI applications.
A Feature Branch (FB) delivers the latest versions of NVIDIA-built and NVIDIA-optimized AI frameworks, NVIDIA NIM microservices, pre-trained models, and SDKs, and provides access to the most recent software features and optimizations.
A Long-Term Support Branch (LTSB) contains long-term supported AI frameworks and SDKs to provide 36 months of API stability and a secure environment for highly regulated industries.
NVIDIA NIM
Part of NVIDIA AI Enterprise, NVIDIA NIM microservices are a set of easy-to-use microservices that accelerate the deployment of foundation models on any cloud or data center and help keep your data secure. NIM microservices have production-grade runtimes, including ongoing security updates. Run your business applications with stable APIs backed by enterprise-grade support.
The support matrices provide a list of supported GPUs, hardware and software capabilities of LLMs, and a list of supported models.
NVIDIA NIM microservices are designed to bridge the gap between the complex world of AI development and the operational needs of enterprise environments, enabling 10-100X more enterprise application developers to contribute to AI transformations of their companies.
The release notes describe the key features, enhancements, and improvements for LLMs and identify known issues.
NVIDIA AI Enterprise Support Lifecycle Policy
This Enterprise Software lifecycle reflects the changes in how NVIDIA AI Enterprise is distributed, which were introduced in NVIDIA AI Enterprise 4.0. Starting with NVIDIA AI Enterprise 4.0, AI frameworks and SDKs are decoupled from software for managing and optimizing infrastructure and workloads.
Notices
What’s Included
The following products are part of NVIDIA AI Enterprise Infrastructure software.
Drivers and Driver Containers
- NVIDIA GPU Data Center Driver Container
- NVIDIA vGPU (C-Series Host and Guest Drivers)
- NVIDIA DOCA-OFED Driver for Networking
Operators
- NVIDIA GPU Operator
- NVIDIA Network Operator
- NVIDIA NIM Operator
- NVIDIA Base Command Manager (BCM)
NVIDIA AI Enterprise Infra Release Branches
Active Infra Release Branches
| Software Branch | Driver Branch | Latest Driver in Branch | Branch | Latest Release in Branch | Latest Release Date in Branch | Branch EOL |
|---|---|---|---|---|---|---|
| NVIDIA AI Enterprise Infra 6.x | R570 | 570.133.12 | Feature and Production | 6.2 | April 2025 | February 2026 |
| NVIDIA AI Enterprise Infra 5.x | R550 | 550.144.02 | Feature and Production | 5.3 | January 2025 | April 2025 |
| NVIDIA AI Enterprise Infra 4.x | R535 | 535.247.01 | Long-Term Support | 4.6 | April 2025 | September 2027 |
All Infra Release Branches
| Software Branch | Driver Branch | Latest Driver in Branch | Branch | Latest Release in Branch | Latest Release Date in Branch | Branch EOL |
|---|---|---|---|---|---|---|
| NVIDIA AI Enterprise Infra 3.x | R525 | 525.147.05 | Feature and Production | 3.3 | November 2023 | December 2023 |
| NVIDIA AI Enterprise Infra 2.x | R520 | 520.61.05 | Feature | 2.3 | October 2022 | November 2022 |
| NVIDIA AI Enterprise Infra 1.x | R470 | 470.256.02 | Long-Term Support | 1.9 | July 2024 | September 2024 |
NVIDIA AI Enterprise Infra Release Documentation
NVIDIA AI Enterprise Infra Release 6.x documentation.
Version 6.2 is the latest release.
Version 6.2 is the latest release.
NVIDIA AI Enterprise Infra Release 5.x documentation.
Version 5.3 is the latest release.
Version 5.3 is the latest release.
NVIDIA AI Enterprise Infra Release 4.x documentation.
Version 4.6 is the latest release.
Version 4.6 is the latest release.
NVIDIA AI Enterprise version 3.x documentation.
NVIDIA AI Enterprise version 2.x documentation.
NVIDIA AI Enterprise version 1.x documentation.
Deployment Guides
This document provides guidance for deploying and running NVIDIA AI Enterprise in the Cloud. This resource can be used for understand system pre-requisites, installation and configuration.
This document provides insights into deploying NVIDIA AI Enterprise for VMware vSphere and serves as a technical resource for understanding system pre-requisites, installation, and configuration.
This document provides insights into deploying NVIDIA AI Enterprise on Bare Metal Servers and serves as a technical resource for understanding system pre-requisites, installation, and configuration.
This document provides insights into deploying NVIDIA AI Enterprise with Red Hat OpenShift on bare metal servers. This technical resource can be used for understanding system pre-requisites, installation, and configuration.
This document provides insights into deploying NVIDIA AI Enterprise with Red Hat OpenShift on VMware vSphere. This technical resource can be used for understanding system pre-requisites, installation, and configuration.
This document provides insights into deploying NVIDIA AI Enterprise on Red Hat Enterprise Linux with KVM Virtualization and serves as a technical resource for understanding system prerequisites, installation, and configuration.
NVIDIA RAPIDS Accelerator for Apache Spark enables data engineers to speed up Apache Spark 3 data science pipelines and AI model training while lowering infrastructure costs.
This guide aims to provide guidance on how to set up a high-performance multi-node cluster as virtual machines. Within this guide, you will become familiar with GPUDirect RDMA and ATS while using Docker as the platform for running high-performance multi-node Deep Learning Training. ATS is a VMware PCIe support enhancement in vSphere 7 Update 2. GPUDirect RDMA benefits from ATS and is certified and supported by NVIDIA AI Enterprise.
This solution guide outlines the creation of an AI pipeline on NVIDIA AI Enterprise by leveraging a Natural Language Processing use case example.
Reference Architecture
This reference architecture provides an example deployment of NVIDIA AI Enterprise software suite. It showcases a deployment with VMWare vShpere, and provides example workloads to showcase the platform’s capabilities. Topics such as hardware, network, and workload topologies will be discussed.
This sizing guide is intended to guide customers who want to implement NVIDIA AI Enterprise with NVIDIA-Certified Systems at scale.
This document outlines a reference architecture for a cost-effective, performant Kubernetes-as-a-Service to build out an infrastructure stack for large-scale AI training and inference workload in virtualized environments.
White Papers
This white paper details NVIDIA's commitment to securing the NVIDIA AI Enterprise software stack. It outlines the processes and measures NVIDIA takes to ensure container security.
This whitepaper outlines the essential components, integration strategies, and tools needed for enterprises to deploy robust, single-tenant AI solutions efficiently and securely with the help of ecosystem partners.
This white paper provides detailed guidance on configuring virtual machines (VMs) to support AI/ML workloads when a hypervisor layer is deployed on top of HGX systems.