Foxconn teams up with Nvidia on AI factory infrastructure development

New Delhi [India], October 16 : Taiwan-based manufacturing giant Foxconn announced that it will partner with US artificial intelligence chip designer Nvidia Corp. to develop AI factory infrastructure, as per a report by Focus Taiwan.

In a statement, Foxconn said at the ongoing OCP Global Summit 2025 in San Jose, California, that it will introduce Nvidia's 800 VDC data center power architecture to its K-1 AI data center project in Kaohsiung for the first time.

The installation will demonstrate Foxconn's efforts in AI servers, data centers and renewable energy integration.

According to Foxconn, the 800 VDC power architecture is designed for high-density AI workloads, featuring a modular and scalable setup that significantly reduces current and resistive losses, minimizes conductor use, simplifies power distribution, and improves energy conversion efficiency and system safety.

The architecture will support next-generation Nvidia's graphics processing units (GPUs), laying a foundation for highly efficient AI servers and data centers with improved power resilience to meet growing computing demand, Foxconn said.

In addition to adopting the new architecture, Foxconn said it will continue to enhance its design capabilities in AI servers, power, and cooling systems to provide rapid solutions for AI data centers.

"Through continuous innovation in high-efficiency power and thermal design, Foxconn reinforces its leadership and highlights Taiwan's pivotal role in the global AI supply chain," the company said.

Also at the OCP Global Summit, Foxconn cloud solutions subsidiary, Ingrasys Technology Inc., showcased several next-generation AI platforms, including the Nvidia GB300 NVL72, Nvidia HGX B300, and Nvidia MGX servers for enterprise-grade AI applications.

Foxconn said these Ingrasys innovations underscore its strong capabilities in AI infrastructure development and system integration.

Disclaimer: This post has been auto-published from an agency feed without any modifications to the text and has not been reviewed by an editor