Nvidia's largest customers have been alerted to significant price increases for servers equipped with the company's artificial intelligence chips. The anticipated hikes, which could surpass 15% in many instances, are driven by soaring costs for memory chips. These new prices are expected to take effect on systems shipped in early next year.
The price adjustments will affect systems featuring Nvidia's high-performance chips, including the Vera Rubin and Grace Blackwell models. The exact percentage of the increase will vary depending on the specific chip generation and memory configurations used in the systems. Companies that manufacture servers under contract for major data center operators, such as Microsoft, Google, and Oracle, have begun notifying their clients about these upcoming changes.
Nvidia, a key player in the buildout of AI infrastructure, is scheduled to release its second-quarter financial results later this week. The company's performance is often viewed as a proxy for the broader AI ecosystem, encompassing both chip manufacturers and businesses involved in the rapid expansion of data center capacity.