(Reuters) -Dell Technologies on Monday unveiled new servers powered by Nvidia’s Blackwell Ultra chips, aiming to capitalize on the booming demand for artificial intelligence systems.
The servers, available in both air-cooled and liquid-cooled variations, support up to 192 Nvidia Blackwell Ultra chips but can be customized to include as many as 256 chips.
Ad Colonial Metals
"This Changes Everything" - Trump Hands Millions Massive IRS Gift
For decades, America's enemies have siphoned off our wealth, hollowed out the middle class, and cheered on the left's wealth-redistributing agenda. And now, these same forces are counting on you to sit back and miss this opportunity to protect your wealth. Which is why we've put together an exclusive, free and brand-new gold guide that shows you how to protect your wealth with gold now.
These servers can train AI models up to four times faster than previous models, Dell said.
The pricing of these products will be “competitive,” Arthur Lewis, President of Dell’s Infrastructure Solutions Group, told Reuters, adding that “there’s a lot of interest on what’s next.”
Dell and Super Micro Computer have benefited from the growing demand for servers designed to handle the computer-heavy AI tasks, but the high cost of producing the systems and tough competition have pressured their margins.
Dell in February forecast a decline in adjusted gross margin rate for fiscal 2026, while Super Micro projected fourth-quarter revenue below estimates earlier this month as tariff-driven economic uncertainty pressures its performance.
Dell will focus on increasing sales of networking and storage products to ensure the “right level of profitability,” Lewis said.
The company’s new servers will also support Nvidia’s upcoming Vera central processing units, which will succeed the chip designer’s Grace server processor.
The AI server maker plans to support Nvidia’s Vera Rubin chips, set to follow the Blackwell series.
Dell also introduced a ‘Pro Max Plus’ laptop designed for AI development on Monday, featuring a neural processing unit that allows engineers to process large AI models directly on the device without relying on cloud services.
(Reporting by Jaspreet Singh in Bengaluru; Editing by Shreya Biswas)
