On October 21, NVIDIA announced the launch of DSX POD, a complete 'AI factory' reference design solution based on the Vera Rubin architecture. The DSX POD is constructed by partners within modular containers, with integration and verification completed prior to delivery. Once the container arrives on-site, it can be immediately operational by connecting to infrastructure such as power, cooling, and networks, reducing the deployment cycle from six months to as fast as 14 days. The DSX POD shifts the integration and installation of computing, networking, cabling, rack-level power, and liquid cooling to manufacturing facilities, reducing the planning, construction, specialized labor, and logistics required for traditional on-site construction. Factory-tested containers mitigate risks associated with component delays and on-site operational errors, shortening the path to model training and production inference, enabling customers to start generating tokens and generating revenue sooner. NVIDIA CEO Jensen Huang stated, 'AI factories create value when they produce tokens, not when they are built. DSX POD brings factory-level precision and speed to AI infrastructure construction, reducing deployment time from months to weeks, allowing customers to scale capacity and launch revenue-generating production faster.'