Nvidia CEO Jensen Huang has confirmed that an upcoming iteration of the company's server family will be liquid cooled.

Huang let slip the detail during a presentation at the 2024 SIEPR Economic Summit at Stanford, but is likely to officially announce the new GPU server system at the company's GTC event from March 18.

"When you look at one of our computers, it's a magnificent thing. It weighs a lot, [has] hundreds of miles of cables," Huang said of the system, potentially a DGX or a different brand.

"The next one - soon coming - is liquid cooled. It's beautiful in lots of ways. And it computes at data center scales."

Earlier this month, Dell's CEO revealed in an earnings call that the upcoming Nvidia B100 GPU would have a thermal design point (TDP) of 1kW - up from 700W in the H100. That would not necessitate direct liquid cooling, he said, but it would definitely be needed “next year with the B200.”

Last year, Nvidia's VP and GM of DGX Systems Charlie Boyle told DCD that the company had prioritized avoiding liquid cooling for its generative AI-focused DGX GH200.

"That was a big system design consideration when talking to our customers," he said.

"We know people will eventually have to move to liquid, but we also hear feedback from customers that that's challenging. For them, their data centers aren't there, they need to build new data centers."

He added: "Even getting liquid cooled equipment - because we're staying ahead of this, building our own stuff internally that does have liquid cooling so that we can test it for our customers - even getting liquid parts, the lead time on it is even much longer.

"One of our core design considerations for this generation was 'how do we still do it on air?'"

Google Cloud, Meta, and Microsoft are among those that have deployed DGX systems in their data centers. The Next Platform last year estimated that DGX machines represent 20 percent of Nvidia's data center revenues.