Newsletter Subscribe
Enter your email address below and subscribe to our newsletter
[forminator_form id="25163"]

finance.yahoo+1finance.yahoo+1finance.biggo+1Nvidia has notified some of its largest customers that prices for servers containing its artificial intelligence chips will rise by more than 15% in many cases, driven by soaring memory chip costs, Bloomberg News reported on Saturday.finance.yahoo+1
The price hikes will take effect on systems shipped early next year and will affect servers built around Nvidia's flagship Vera Rubin and Grace Blackwell platforms, according to people familiar with the process cited by Bloomberg. The magnitude of the increases will vary depending on chip generation and memory configuration.finimize+1
Companies that build servers under contract for large data center operators such as Microsoft , Alphabet's Google, and Oracle have recently informed their customers of the upcoming increases, according to the report. The primary driver is the surging cost of high-bandwidth memory and DRAM, critical components in AI accelerators that have grown more expensive as suppliers struggle to keep pace with demand.finance.biggo+2
Pre-shipment quotes for the NVL72 VR200 system based on the Vera Rubin architecture have already reached $5 million to $7 million per unit, while earlier Blackwell-based NVL72 GB200 racks were quoted between $2.8 million and $3.4 million.moomoo
The news arrives days before Nvidia is set to report fiscal second-quarter results on August 26, with Wall Street expecting revenue to roughly double from a year earlier. In its most recent quarter ended in April, the company posted revenue of $81.6 billion while maintaining gross margins around 75%.finance.yahoo+2
Nvidia controls more than 70% of the data center AI chip market and commands the world's largest market capitalization. The price increases could accelerate efforts by major customers including Amazon Amazon.com, Inc. , Microsoft, Google, and Meta to develop proprietary chip alternatives, though Nvidia's software ecosystem continues to make switching difficult.moomoo+1
AMD Advanced Micro Devices, Inc. has positioned its MI300 series as a more cost-effective alternative, while cloud providers have invested in custom silicon such as Amazon Web Services' Trainium chips. Still, most newly built large-scale data centers remain heavily reliant on Nvidia hardware.techbuzz+1
Reuters said it could not immediately verify the Bloomberg report, and Nvidia did not respond to requests for comment outside regular business hours.finance.yahoo