Archive · 23 Aug 2026
Nvidia Customers Face Reported AI Server Price Hikes Above 15%
Nvidia makes chips that help computers run artificial intelligence.
Reports say some servers using those chips may become more than 15% more expensive.
The higher prices are expected for servers shipped early next year.
The exact increase will depend on the Nvidia chip and the amount of memory in the server.
Memory chips are becoming more expensive because AI data centers need so many of them.
Samsung Electronics, SK hynix and Micron Technology supply much of the world’s DRAM memory.
Server companies have reportedly warned customers such as Microsoft, Google and Oracle about higher costs.
Nvidia has not confirmed the reported changes, and Reuters said it could not independently verify them.
Higher prices could make it harder and more expensive for companies to build AI data centers.
Some major Nvidia customers were told AI-server prices could rise by more than 15% in many cases.
The increases are expected for systems shipped early next year, including Vera Rubin and Grace Blackwell configurations.
Final price changes will vary according to Nvidia’s chip generation and memory configuration.
Contract manufacturers serving Microsoft, Google and Oracle have reportedly notified customers of higher prices.
Soaring memory costs could raise data-center expenses and complicate major companies’ AI expansion plans.
- Who
- Nvidia, its major customers, contract server manufacturers, and memory suppliers including Samsung Electronics, SK hynix and Micron Technology.
- What
- Nvidia reportedly warned that prices for some AI servers could increase by more than 15%.
- When
- The increases are expected to apply to systems shipped early next year; the reports were published on August 23, 2026.
- Where
- The reported increases affect AI-server supply chains serving data-center operators including Microsoft, Google and Oracle.
- Why
- Rising memory-chip costs and strong demand for AI infrastructure are creating pricing pressure.
Supply-Chain Pressures
Customer and Industry Concerns
Cause of higher prices
Supply-Chain Pressures
The reported increases reflect soaring memory costs, constrained component supply and strong demand for AI infrastructure.
Customer and Industry Concerns
Customers and data-center operators may face substantially higher costs for servers and expansion projects.
Who absorbs the cost
Supply-Chain Pressures
Reports suggest Nvidia may pass much of the higher memory expense to customers rather than absorb it entirely, although the exact split is unclear.
Customer and Industry Concerns
Server assemblers, cloud providers and other companies may need to decide whether to absorb the increases or pass them on to their own customers.
Potential alternatives
Supply-Chain Pressures
Nvidia remains central to many AI data-center build-outs, while rival systems may face the same memory-supply constraints.
Customer and Industry Concerns
Amazon, Microsoft, Google and Meta are pursuing in-house AI-chip programs that could eventually reduce their reliance on Nvidia, depending on access to memory supplies.
- Reported increase
- More than 15% for some AI servers
- Expected timing
- Systems shipped early next year, potentially in early 2027
- Affected platforms
- Systems including Nvidia’s Vera Rubin and Grace Blackwell families
- Pricing factors
- Nvidia chip generation and memory configuration
- Reported memory suppliers
- Samsung Electronics, SK hynix and Micron Technology
- Nvidia gross margin
- 75%, according to the reports
- Verification status
- Nvidia has not confirmed the changes; Reuters said it could not immediately verify the report
Sources
Nvidia customers notified about AI-related price hikes above 15%
Nvidia AI servers may get 15% costlier — What it signals for the industry
Nvidia reportedly warns some of its biggest customers about upcoming price hike of AI servers by 15%: Details