NVIDIA Unveils DSX POD to Accelerate European AI Infrastructure Deployment

NVIDIA introduced DSX POD at GTC Berlin, a reference design for complete NVIDIA Vera Rubin AI factories built by partners inside modular containers and integrated and validated before delivery. The factory-built units arrive ready to connect to a prepared site’s power, cooling, networking and other supporting infrastructure, reducing deployment time from about six months to as little as 14 days. The design moves installation of computing, networking, cabling, rack-level power and liquid cooling to the manufacturing facility, reducing the planning, construction, specialized labor and logistics required for traditional onsite builds. By shifting assembly and validation to controlled factory environments, the approach lowers exposure to component delays and costly onsite errors, allowing customers to move more quickly to model training and production inference. NVIDIA said the goal is to shorten the path from delivery to the first production tokens, when an AI factory begins serving production workloads and generating revenue. NVIDIA founder and CEO Jensen Huang said AI factories create value when they start producing tokens, not when construction is complete, adding that DSX POD brings factory-built precision and speed to AI infrastructure deployment. NVIDIA is developing DSX POD with a global partner ecosystem including Anduril, Armada, Bull, Crusoe, DayOne, Dell Technologies, Delta, Eaton, Foxconn, GIGABYTE, HPE, Lenovo, Pegatron, Quanta Cloud Technology, Schneider Electric, Vertiv and Wistron. The company said the factory-built AI infrastructure units are expected to be available worldwide in early 2027.