SANTA CLARA, CALIFORNIA / RankWire.AI / – Nvidia is planning to increase prices by more than 15% for numerous AI server configurations expected to ship in early 2027. These price hikes affect setups using the company’s Vera Rubin and Grace Blackwell platforms, with the exact increase varying depending on the chip generation and memory configuration. Nvidia has not issued a broad price hike across all server offerings publicly, but server manufacturers have communicated the new pricing details to leading data center clients.
Major data center operators such as Microsoft, Google, and Oracle have been informed of these price changes by server manufacturers. These companies purchase significant volumes of AI infrastructure for cloud and data center services. The escalating costs of memory have become a key factor influencing the reported increases in server prices. AI applications demand large quantities of high-bandwidth and server memory in addition to GPUs, CPUs, networking hardware, and storage, which has kept certain memory categories in short supply throughout 2026.
According to TrendForce, conventional DRAM contract prices are expected to rise between 13% and 18% in the third quarter of 2026. Simultaneously, NAND Flash contract prices are forecasted to increase by 10% to 15% during the same period. The research firm pointed out that server DRAM remains undersupplied as suppliers prioritize capacity for AI-related applications. While some long-term supply agreements have limited the extent of price increases, overall costs for server memory continue to be high.
Memory prices climb amid persistent high demand for AI servers
In May, Nvidia announced that Vera Rubin had reached full production with server manufacturers and supply-chain partners worldwide. The company stated that products incorporating Rubin would be available in the second half of 2026. Vera Rubin combines the Vera CPU, Rubin GPU, NVLink 6, ConnectX-9, BlueField-4, and Spectrum-6 networking technologies. This platform replaces Grace Blackwell as Nvidia’s latest rack-scale architecture designed for large AI workloads. System manufacturers are developing Rubin-based products for cloud providers, AI research labs, and hyperscale data centers.
Grace Blackwell remains a key part of Nvidia’s current data center lineup. The GB200 NVL72 design connects 36 Grace CPUs with 72 Blackwell GPUs within a single liquid-cooled rack. Nvidia built this system to function as a unified NVLink computing domain suitable for large AI models. The reported price increases for servers vary depending on configuration, with memory capacity and chip generation being significant factors. Consequently, the 15% increase does not apply uniformly to all Nvidia server configurations.
Expansion of Vera Rubin production impacts Nvidia’s server ecosystem
Memory suppliers have shifted production toward server and high-performance products as AI demand continues to consume capacity. TrendForce indicated that this shift has reduced supply for certain PC and consumer DRAM products. Despite strong server shipments throughout 2026, buyers have maintained building inventories of data center memory. The market researcher predicts that server DRAM supply will remain tight into 2027, as demand growth surpasses new supply, creating the backdrop for Nvidia’s reported price adjustments.
Nvidia begins this pricing phase with record data center sales. In its fiscal first quarter ending April 26, the company posted total revenue of $81.6 billion. Data Center revenue hit a record $75.2 billion, representing a 92% increase compared to the previous year. Nvidia projected second-quarter revenue at $91 billion, plus or minus 2%, when it announced these results in May. The company is scheduled to release its fiscal second-quarter results on Aug. 26.