Setting up Artificial Intelligence factories in simulation, decreasing deployment time from months to days, is accelerating the next industrial revolution. At GTC 2026 in San Jose, NVIDIA introduced DSX Air as part of NVIDIA DSX Sim in the DSX platform, a blueprint for Artificial Intelligence factories. The software-as-a-service platform delivers high-fidelity digital simulations of NVIDIA hardware infrastructure, including GPUs, SuperNICs, DPUs and switches, and connects with partner technologies for storage, routing, security and orchestration through open, API-based integration.
DSX Air is designed to let organizations build a full digital twin of an Artificial Intelligence factory across compute, networking, storage, orchestration and security before physical systems arrive. By moving integration and troubleshooting into simulation, customers are reducing the time to first token from weeks or months to mere days or hours, while lowering cost and operational risk. NVIDIA frames the platform as both a deployment accelerator and an operating model for managing change over time, allowing teams to validate upgrades, patches and maintenance plans in simulation before applying them to production.
The platform is intended to support a broad ecosystem of server manufacturers, orchestration vendors, storage providers and security partners in a shared environment. NVIDIA highlighted demonstrations with Netris for network orchestration, Rafay for host orchestration and NVIDIA Run:ai for GPU allocation in a multi-tenant RTX PRO Server setup running in simulation. A separate demo used the VAST AI Operating System for a video retrieval-augmented generation workflow, including a VAST cluster with DataEngine nodes and a video search and summarization front end. Security testing was shown with Check Point distributed firewall software on simulated BlueField DPUs, TrendAI Vision One for threat detection and Keysight Cyperf generating traffic.
NVIDIA said companies including CoreWeave are already using DSX Air to simulate and validate environments before hardware reaches the loading dock. Siam.AI, Thailand’s largest Artificial Intelligence cloud provider, is using simulation to adopt NVIDIA best practices ahead of hardware delivery and validate architecture in a virtual environment. Hydra Host is using DSX Air to accelerate development of Brokkr, its Artificial Intelligence factory operating system for bare-metal GPU provisioning that is used by dozens of GPU deployments globally. NVIDIA positions simulation as a foundational layer for deploying and operating increasingly complex Artificial Intelligence infrastructure at scale.
