Is this page helpful?

Quick Start Guide#

In about 30–60 minutes you can activate your enterprise account, install NVIDIA AI Enterprise software, and confirm GPU containers run. Follow the steps for your path: bare metal (Ubuntu + Docker), VMware vSphere, or public cloud (Azure walkthrough below).

What You Will Accomplish

When you finish, you will have:

Activated your NVIDIA Enterprise Account and accessed the NVIDIA NGC Catalog
Installed NVIDIA AI Enterprise software components
Verified GPU-accelerated containers are working

Next Steps

After this guide, use Where to Go Next for license tasks, software catalogs, and deeper deployment documentation.

Prerequisites#

Before you start, confirm you have:

Hardware Requirements

NVIDIA-Certified Server Platform that supports NVIDIA AI Enterprise
One or more supported NVIDIA GPUs installed in your server

Important

The NVIDIA-Certified Systems requirement does not apply to GB200 NVL4, GB200 NVL72, and GB300 NVL72 systems. For these platforms, NVIDIA-Qualified server status is the prerequisite for NVIDIA AI Enterprise support. For more information, refer to the NVIDIA Qualified Systems Catalog.

Software & Licensing

Valid NVIDIA AI Enterprise software subscription
For bundled GPUs (such as NVIDIA H100 PCIe), activate your license before installation

Additional Information

Refer to the NVIDIA AI Enterprise Release Notes for supported hardware, version matrix, and known issues.

Attention

Already have an account? Skip directly to Installing NVIDIA AI Enterprise Software Components.

Note

The following instructions do not apply to NVIDIA DGX systems. Refer to NVIDIA DGX Systems for DGX-specific documentation.

Activating the Accounts for NVIDIA AI Enterprise#

Your Order Confirmation Message#

After your order is processed, you get an order confirmation with an attached NVIDIA Entitlement Certificate (product activation keys and usage steps).

If you are a data center administrator, follow the certificate instructions. If not, forward the full email and attachment to your data center administrator.

What Your NVIDIA Enterprise Account Provides#

Sign in at the NVIDIA Application Hub to reach:

NVIDIA NGC — software catalog, services, and management tools under your subscription
NVIDIA Enterprise Support Portal — support cases and downloads for NVIDIA AI Enterprise
NVIDIA Licensing Portal — entitlements and license server setup (required before NVIDIA vGPU for Compute)

Creating your NVIDIA Enterprise Account#

Prerequisites

Have your order confirmation message with the NVIDIA Entitlement Certificate ready
Choose a unique email address (if creating a new account separate from an evaluation account)

Steps

Follow the Register link in the instructions for using your NVIDIA Entitlement Certificate.
Fill out the NVIDIA Enterprise Account Registration page form and click REGISTER. A message confirming that an account has been created will appear. The email address you provided will then receive an email instructing you to log in to your account on the NVIDIA Application Hub.
Open the email instructing you to log in to your account and click Log In.
On the open NVIDIA Application Hub Login page, type the email address you provided in the text-entry field and click Sign In.
On the Create Your Account page that opens, provide and confirm a password for the account and click Create Account. A message prompting you to verify your email address appears. An email instructing you to verify your email address is sent to your provided email address.
Open the email instructing you to verify your email address and click Verify Email Address. A message confirming that your email address is confirmed appears.

Your account is now active. Continue to Installing NVIDIA AI Enterprise Software Components.

Linking an Evaluation Account to an NVIDIA Enterprise Account for Purchased Licenses#

Link your evaluation account to purchased licenses by registering with the same email address you used for your evaluation account. To create a separate account instead, see Creating your NVIDIA Enterprise Account and use a different email address.

Follow the Register link in the instructions for using the NVIDIA Entitlement Certificate for your purchased licenses.
Fill out the NVIDIA Enterprise Account Registration page form, specifying the email address with which you created your existing account, and click Register.
When a message stating that your email address is already linked to an evaluation account is displayed, click LINK TO NEW ACCOUNT.
Log in to the NVIDIA Licensing Portal with the credentials for your existing account.

Installing NVIDIA AI Enterprise Software Components#

Get NVIDIA AI Enterprise components from the NVIDIA NGC Catalog. Main entry points:

NVIDIA AI Enterprise Software Suite — frameworks, NIM microservices, models, and application tooling
NVIDIA AI Enterprise Infra 8 collection — GPU Operator, Network Operator, device plugins, and related infra
NVIDIA Omniverse — OpenUSD-based apps and workflows (industrial digitalization, simulation, physical AI)
NVIDIA Run:ai collection — self-hosted GPU scheduling and workload orchestration

Sign in to NVIDIA NGC from the NVIDIA NGC Sign In page before you download assets.

Application Software

Data scientists, ML engineers, and developers use the application layer to build and ship models and apps on the GPU platform. The Software Suite bundles AI frameworks, NIM microservices, and pretrained models on NGC.

Go to NVIDIA AI Enterprise Supported on NVIDIA NGC to view the NVIDIA AI Enterprise Software Suite.
Browse the NVIDIA AI Enterprise Software Suite to find software assets.
Click an asset to learn more about it or download it.

Infrastructure Software

Cluster administrators and IT operations use the infrastructure layer to install and run the GPU platform. The NVIDIA AI Enterprise Infra collection includes GPU Operator, Network Operator, device plugins, and related orchestration software.

Go to the NVIDIA AI Enterprise Infra 8 collection on NVIDIA NGC.
Click the Artifacts tab and select the resource.
Click Download and choose to download the resource using a direct download in the browser, the displayed wget command, or the NVIDIA NGC CLI.

Omniverse

NVIDIA Omniverse is included with your NVIDIA AI Enterprise license for OpenUSD-based build and deploy workflows (industrial digitalization, simulation, physical AI).

Go to the NVIDIA Omniverse page on NVIDIA NGC.
Browse the available components, including Omniverse Kit SDK, Nucleus, and enterprise applications.
Click an asset to learn more about it or download it.

For more information about NVIDIA Omniverse, refer to the NVIDIA Omniverse documentation.

Run:ai

NVIDIA Run:ai is GPU scheduling and AI workload orchestration for clusters.

Note

The NVIDIA AI Enterprise license includes NVIDIA Run:ai self-hosted deployments only. Run:ai SaaS is not included in the NVIDIA AI Enterprise license and remains a separate offering.

Self-hosted Run:ai lives in its own NGC collection (not inside the Infra collection).

Go to the NVIDIA Run:ai Self-Hosted collection on NVIDIA NGC.
Follow the steps in the NVIDIA Run:ai Self-Hosted Installation Guide.

Accessing Container Images from NGC#

NVIDIA AI Enterprise ships as container images on the NVIDIA NGC Catalog. A typical image bundles the runtime stack (CUDA, cuDNN, Magnum IO, TensorRT, and the target framework).

Before you pull images, complete:

From the host machine, obtain the Docker pull command from the component listing in the NGC Catalog.

Refer to the NVIDIA AI Enterprise NGC supported software for the full image list.

Choose Your Deployment Type#

Pick the row that matches your environment, then open the linked procedure.

Table 2 Deployment Types#
Category	Description	Getting Started
Bare Metal	Direct installation on physical servers with dedicated GPUs	Installing NVIDIA AI Enterprise on Bare Metal Ubuntu Obtaining NVIDIA Base Command Manager
Virtualized	Deployment on VMware vSphere using NVIDIA vGPU for Compute	Installing NVIDIA AI Enterprise on VMware vSphere
Public Cloud	Deployment on cloud platforms (AWS, Azure, GCP, OCI, Alibaba, Tencent)	Installing NVIDIA AI Enterprise on Microsoft Azure

Note

For prerequisites, install, configuration, and advanced scenarios beyond this quick start, refer to Planning & Deployment on the NVIDIA AI Enterprise Docs Hub.

Installing NVIDIA AI Enterprise on Bare Metal Ubuntu#

Deploy a single-node bare metal stack with Docker on NVIDIA-Certified Systems (or NVIDIA-Qualified Systems for GB200 NVL4, GB200 NVL72, and GB300 NVL72).

Prerequisites#

Install the NVIDIA data center GPU driver:

Download from NVIDIA Drivers.
Within the Manual Driver Search section, select Data Center/Tesla, your architecture type, and Linux 64-bit to download the .run file.

Installing NVIDIA AI Enterprise using the TRD Driver#

The NVIDIA AI Enterprise Linux driver install needs a compiler toolchain and kernel headers.

Prerequisites

Complete the CUDA pre-installation steps first.

Note

If you prefer to use a Debian package to install CUDA, refer to the Debian instructions.

Steps

Log into the system and check for updates.
```
sudo apt-get update
```
Install the GCC compiler and the make tool in the terminal.
```
sudo apt-get install build-essential
```
Copy the NVIDIA AI Enterprise Linux driver package, for example, NVIDIA-Linux-x86_64-xxx.xx.xx.run, to the host machine where you are installing the driver.

Where xxx.xx.xx is the current NVIDIA AI Enterprise version and driver version.
Navigate to the directory containing the NVIDIA Driver .run file. Then, add the Executable permission to the NVIDIA Driver file using the chmod command.
```
sudo chmod +x NVIDIA-Linux-x86_64-xxx.xx.xx.run
```
From a console shell, run the driver installer as the root user and accept the defaults.
```
sudo sh ./NVIDIA-Linux-x86_64-xxx.xx.xx.run
```
Reboot the system.
```
sudo reboot
```
After the system has rebooted, confirm that you can see your NVIDIA GPU device in the output from nvidia-smi.
```
# Verify GPU driver installation and GPU detection
$ nvidia-smi
```
Expected output: GPU information table showing driver version, CUDA version, and GPU name. If no output appears or you see "command not found", confirm the driver install finished and reboot if needed.

Installing the NVIDIA Container Toolkit#

The NVIDIA Container Toolkit adds libnvidia-container and runtime hooks so Docker can use NVIDIA GPUs. Refer to the NVIDIA Container Toolkit documentation for architecture details.

Installation Steps

Install Docker — Install Docker Engine on Ubuntu (Install Docker Engine on Ubuntu).
Install the NVIDIA Container Toolkit — Add the repo and packages (Installing the NVIDIA Container Toolkit).
Configure the runtime — Point Docker at the NVIDIA Container Runtime (Configuration).

Verification Steps

Run the nvidia-smi command from a CUDA container image. Use a CUDA Toolkit image version that matches or is earlier than your installed driver version.

Note

Use a CUDA Toolkit image version that does not exceed your driver version. For example, if your driver supports CUDA 12.x, use a CUDA 12.x or earlier container image. For a list of available images, refer to nvidia/cuda on Docker Hub.
```
# Test GPU access from container (uses all available GPUs)
# --rm: automatically remove container after exit
# --runtime=nvidia: use NVIDIA Container Runtime
# --gpus all: expose all GPUs to container
$ sudo docker run --rm --runtime=nvidia --gpus all nvidia/cuda:12.8.1-base-ubuntu22.04 nvidia-smi
```
Troubleshooting
icon:

wrench

If you encounter nvidia-container-cli: initialization error, verify:
1. NVIDIA driver is loaded: nvidia-smi returns successfully on host
2. Container toolkit is installed: which nvidia-container-cli
3. Docker daemon restarted: sudo systemctl restart docker

Start a GPU-enabled container on any two available GPUs.

$ docker run --runtime=nvidia --gpus 2 nvidia/cuda:12.8.1-base-ubuntu22.04 nvidia-smi

Start a GPU-enabled container on two specific GPUs identified by their index numbers.

$ docker run --runtime=nvidia --gpus '"device=1,2"' nvidia/cuda:12.8.1-base-ubuntu22.04 nvidia-smi

Start a GPU-enabled container on two specific GPUs, with one GPU identified by its UUID and the other GPU identified by its index number.
```
$ docker run --runtime=nvidia --gpus '"device=UUID-ABCDEF,1"' nvidia/cuda:12.8.1-base-ubuntu22.04 nvidia-smi
```

Specify a GPU capability for the container.

$ docker run --runtime=nvidia --gpus all,capabilities=utility nvidia/cuda:12.8.1-base-ubuntu22.04 nvidia-smi

Obtaining NVIDIA Base Command Manager#

NVIDIA Base Command Manager handles cluster provisioning, workload management, and infra monitoring on bare metal.

Activate your NVIDIA Enterprise Account (Activating the Accounts for NVIDIA AI Enterprise).
Email your entitlement certificate to sw-bright-sales-ops@NVIDIA.onmicrosoft.com to request your NVIDIA Base Command Manager product keys. After your entitlement certificate has been reviewed, you will receive a product key. Generate a license key for the number of licenses you purchased.
Download NVIDIA Base Command Manager for your operating system.
Follow the steps in the NVIDIA Base Command Manager Installation Manual to create and license your head node.

For more information, refer to the Base Command Manager product manuals.

Installing NVIDIA AI Enterprise on VMware vSphere#

Refer to the NVIDIA AI Enterprise VMware Deployment Guide to deploy on a third-party NVIDIA-Certified System with VMware vSphere and NVIDIA vGPU for Compute.

Prerequisites#

NVIDIA Virtual GPU Manager
NVIDIA vGPU for Compute Guest Driver
NVIDIA License System

To download the NVIDIA vGPU for Compute software drivers, follow the instructions for Accessing the NVIDIA AI Enterprise Infrastructure Software.

Your NVIDIA AI Enterprise subscription covers downloading vGPU for Compute software from NGC; use Installing NVIDIA AI Enterprise Software Components to sign in and browse. vGPU deployments need the NVIDIA License System to enforce entitlements.

Installing NVIDIA AI Enterprise on Microsoft Azure#

NVIDIA AI Enterprise runs on AWS, Google Cloud, Microsoft Azure, OCI, Alibaba Cloud, and Tencent Cloud.

This section is a short Azure path using the NVIDIA AI Enterprise VMI. For other clouds, start from Planning & Deployment on the Docs Hub.

Installing NVIDIA AI Enterprise on Microsoft Azure using the NVIDIA AI Enterprise VMI#

The On-Demand VMI provisions a GPU VM with common AI/HPC tooling already installed.

Preinstalled software includes:

Ubuntu Operating System
NVIDIA GPU Data Center Driver
Docker-ce
NVIDIA Container Toolkit
CSP CLI, NGC CLI
Miniconda, JupyterLab, Git
Token Activation Script

On the Enterprise (On-Demand) VMI, do two things:

Authorize the instance with NGC using the instance ID token on the Activate Subscription page. Work through these four sub-steps:
1. Get an identity token from the VMI.
2. Activate your NVIDIA AI Enterprise subscription with the token.
3. Generate an API key to access the catalog.
4. Put the API key on the VMI.
Follow the NGC Catalog access instructions to complete this step.
Pull and run NVIDIA AI Enterprise containers (pull and run NGC images).

For a more full public-cloud walkthrough, refer to the Microsoft Azure Overview.

Deployment Types#

Use these links when you need full deployment docs per environment. Match the row to your infrastructure.

Table 3 Deployment Options#
Type	Description
Bare Metal	Direct installation on physical servers with dedicated GPUs
Virtualized	Deployment on VMware vSphere using NVIDIA vGPU for Compute or GPU Passthrough
Public Cloud	Deployment on cloud platforms (AWS, Azure, GCP, OCI, Alibaba, Tencent)

For prerequisites, install, configuration, and advanced scenarios, refer to Planning & Deployment on the NVIDIA AI Enterprise Docs Hub.

Where to Go Next#

After this quick start, use these resources as needed:

Table 4 Where to Go Next#
If you need to…	Go to	Why
Start: Get started from scratch (account, drivers, first workload)	This Document	Guides you through account activation, software installation, and running your first AI workload
Discover: Browse available software components	NVIDIA AI Enterprise Software	Complete catalog of Application and Infrastructure Layer components with links to NGC Catalog entries
Research: Choose a release branch or check version compatibility	NVIDIA AI Enterprise Lifecycle Policy	Defines branch types (FB, PB, LTSB, Infrastructure), support periods, and includes the Interactive Lifecycle and Compatibility Explorer
Plan: Plan a deployment architecture	Planning & Deployment on the NVIDIA AI Enterprise Docs Hub	Reference architectures, sizing guides, and deployment blueprints
Validate: Validate your infrastructure stack	Interactive Lifecycle and Compatibility Explorer	Confirm that your NVIDIA GPU driver, NVIDIA GPU Operator, NVIDIA Network Operator, and NVIDIA Run:ai versions are compatible before deploying
Application layer: Deploy or configure NVIDIA Omniverse	NVIDIA Omniverse Documentation	Standalone documentation site — NVIDIA Omniverse is included in your NVIDIA AI Enterprise license, but is documented separately
Kubernetes operators: Deploy NVIDIA GPU Operator, NVIDIA Network Operator, or NVIDIA DPU Operator (DPF)	NVIDIA GPU Operator Documentation · NVIDIA Network Operator Documentation · NVIDIA DOCA Platform Framework (DPF) Documentation	Standalone documentation sites for Kubernetes operator installation and configuration
AI frameworks and NIMs: Set up NVIDIA NIM, NVIDIA NeMo, or other AI frameworks	NVIDIA NIM Documentation · NVIDIA NeMo Documentation · NVIDIA NGC Catalog	Standalone documentation sites; the NVIDIA NGC Catalog lists all supported software
Infrastructure orchestration: Deploy NVIDIA Run:ai	NVIDIA Run:ai Documentation	Standalone documentation site — NVIDIA Run:ai self-hosted is included in your NVIDIA AI Enterprise license (NVIDIA Run:ai SaaS is separate)
Support: Check licensing or open a support case	Support on the NVIDIA AI Enterprise Docs Hub	NVIDIA Enterprise Support Services portal for licensing and technical support