Nvidia AI Server Prices Rising Over 15% on Memory Costs

Price increases target flagship AI systems

Nvidia has informed its largest customers that prices for servers equipped with its AI chips will increase by more than 15% in many configurations, according to a Bloomberg News report published Saturday. The price adjustments stem from soaring memory chip costs and will take effect on systems shipped in early 2025.

The increases will affect servers containing Nvidia's flagship Vera Rubin and Grace Blackwell chip architectures, according to people familiar with the notifications. The exact percentage increase will vary based on chip generation and memory configuration choices.

Contract manufacturers that assemble servers for major data center operators—including Microsoft, Alphabet's Google, and Oracle—have begun notifying their customers of the coming price changes, Bloomberg reported. These server builders act as intermediaries between Nvidia's chip supply and the hyperscale cloud providers driving AI infrastructure expansion.

Why it matters

The price hikes arrive as enterprises face mounting pressure to justify massive AI infrastructure investments. For cloud providers already spending billions on data center buildouts, a 15% increase on server costs compounds capital expenditure challenges at a time when AI monetization remains uncertain for many applications. The timing also underscores how memory bottlenecks—not just GPU availability—now constrain AI deployment economics.

Timing ahead of earnings

Nvidia is scheduled to report its second-quarter financial results on August 26. The company has become a bellwether for the broader AI ecosystem, with its performance closely watched by investors tracking both chip makers and the companies financing rapid data center capacity expansion.

Reuters could not independently verify the Bloomberg report, and Nvidia did not respond to requests for comment outside regular business hours.

The price increases reflect broader supply chain pressures in the semiconductor industry, where high-bandwidth memory—essential for AI workloads—has experienced tight supply and rising costs. These specialized memory components enable the massive data throughput required for training and running large language models and other AI applications.

Nvidia's chips currently underpin much of the AI infrastructure buildout across the technology industry, giving the company significant pricing power even as competition intensifies from rivals including AMD and custom silicon efforts from cloud providers themselves.

The details were first reported by Bloomberg News.