Infineon and NVIDIA Collaborate on 800V Power Architecture for AI Data Centers

May 21, 2025
Infineon Technologies is partnering with NVIDIA to develop an industry-first 800V high-voltage direct current (HVDC) power delivery architecture for AI data centers, announced in a press release. This new system aims to enhance energy efficiency and reliability in AI server racks.

Infineon Technologies is collaborating with NVIDIA to develop an industry-first 800V high-voltage direct current (HVDC) power delivery architecture for AI data centers, announced in a press release. This new architecture is designed to significantly improve energy-efficient power distribution across data centers and enable direct power conversion at the AI chip level within server boards.

The collaboration aims to set new standards for power delivery in AI data centers, which are expected to require power outputs of one megawatt (MW) or more per IT rack by the end of the decade. The centralized HVDC system will replace the current decentralized power supply approach, optimizing space and reducing the number of power conversion stages needed.

Infineon's expertise in power conversion solutions, utilizing semiconductor materials such as silicon (Si), silicon carbide (SiC), and gallium nitride (GaN), is key to accelerating the roadmap towards a full-scale HVDC architecture. This initiative is expected to enhance the reliability and efficiency of AI infrastructure, supporting the growing demands of AI workloads.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.