NVIDIA AI Enterprise Cloud Deployment

Type: Deployment Guide Tags: NVIDIA, AI Enterprise, cloud, AWS, Azure, Google Cloud, OCI, Alibaba Cloud, Tencent Cloud, Kubernetes, VMI Related: NVIDIA-AI-Enterprise, NVIDIA-AI-Enterprise-Lifecycle-Policy, NVIDIA-Enterprise-Licensing-Guide, NVIDIA-DGX-Cloud, NVIDIA-Cloud-Accelerator-NCX, NVIDIA-NIM, NVIDIA-NIM-on-GKE, NVIDIA-NGC-Catalog, NGC, NVIDIA-Container-Toolkit, NVIDIA-GPU-Operator, NVIDIA-vGPU, NVIDIA-AI-Enterprise-Bare-Metal-Deployment, NVIDIA-AI-Enterprise-VMware-Deployment, Red-Hat-AI-Factory-with-NVIDIA Sources: https://docs.nvidia.com/ai-enterprise/deployment/cloud/latest/overview.html Last Updated: 2026-04-29

Summary

NVIDIA AI Enterprise Cloud Deployment is the current NVIDIA guide for running AI Enterprise on supported cloud service providers. It covers AWS, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, Alibaba Cloud, Tencent Cloud, ByteDance Volcano Engine Cloud, NVIDIA VMIs, GPU-optimized VMIs, standard instances, managed Kubernetes, Red Hat OpenShift in the cloud, private offers, licensing, and NGC container access.

Detail

Purpose

Cloud deployment lets organizations run AI Enterprise with public-cloud GPU capacity while keeping NVIDIA’s supported software stack, licensing model, and deployment guidance. The guide is the cloud counterpart to the bare-metal and VMware deployment guides.

Deployment scope

  • Supported CSP sections for AWS, Google Cloud, Azure, OCI, Alibaba Cloud, Tencent Cloud, and ByteDance Volcano Engine Cloud.
  • Standard GPU instances with customer-managed setup.
  • NVIDIA Private Offer VMIs and NVIDIA AI Enterprise VMIs with license-included paths.
  • NVIDIA GPU-Optimized VMIs for preconfigured GPU software without included AI Enterprise licensing.
  • Managed Kubernetes paths for AWS, Azure, and Google Cloud.
  • Marketplace Red Hat OpenShift in AWS, Azure, and Google Cloud.
  • NGC container pull/run guidance.
  • BYOL, on-demand/pay-as-you-go, and private-offer licensing notes; the current NVIDIA-Enterprise-Licensing-Guide also ties BYOL consumption to per-GPU subscription licensing on certified cloud providers.

NVIDIA context

Use this page for AI Enterprise in public clouds. Use NVIDIA-DGX-Cloud for NVIDIA-hosted/partner AI supercomputing capacity and NVIDIA-Cloud-Accelerator-NCX for cloud partner AI infrastructure components.

Connections

Source Excerpts

  • The cloud guide lists AWS, Google Cloud, Microsoft Azure, OCI, Alibaba Cloud, and Tencent Cloud as AI Enterprise cloud targets.
  • Deployment methods include NVIDIA VMIs, private offer VMIs, GPU-optimized VMIs, managed Kubernetes, and marketplace Red Hat OpenShift.

Resources