Nvidia unveils DSX POD modular AI factory design for 2027 delivery

Nvidia (NASDAQ: NVDA) announced DSX POD, a reference design for containerized AI factories based on its Vera Rubin systems, at GTC Berlin on Oct. 21, 2026. The design moves rack integration, cabling, cooling and functional validation to the manufacturing facility before delivery, rather than performing that work onsite.

According to the company, the approach reduces deployment time at a prepared site from approximately six months to as little as 14 days. The prepared site must supply power, cooling, networking and other supporting infrastructure before the units arrive.

"An AI factory creates value when it starts producing tokens — not when construction is complete," said Jensen Huang, founder and CEO of Nvidia. "With NVIDIA DSX POD, we're bringing factory-built precision and speed to AI infrastructure construction — reducing deployment time from months to weeks so customers can scale capacity and bring revenue-generating production online faster."

Nvidia said it is working with a group of computing and infrastructure partners to manufacture and deliver DSX POD units worldwide in early 2027. Partners named in the announcement include Anduril, Armada, Bull, Crusoe, DayOne, Dell Technologies, Delta, Eaton, Foxconn (Ingrasys), Gigabyte, HPE, Lenovo, Pegatron, Quanta Cloud Technology, Schneider Electric, Vertiv and Wistron.

The DSX POD is part of Nvidia's broader DSX infrastructure platform. Partners complete rack-to-rack cabling, wiring, piping and functional validation before shipment, with site preparation occurring separately and in parallel.

You May Also Be Interested In

• JPMorgan says tech stocks lead U.S. futures lower on risks
• Cheniere signs 30-year LNG supply deal with QatarEnergy
• Stagwell and Palantir expand AI marketing partnership