SANTA CLARA, CALIFORNIA / RankWire.AI / – Nvidia is planning to increase prices by more than 15% on many of its AI server configurations scheduled for shipment in early 2027. These price hikes affect setups built on the company’s Vera Rubin and Grace Blackwell platforms. The extent of each increase varies depending on the chip generation and memory configuration. Nvidia has not officially disclosed a uniform price hike across all its server offerings. Additionally, server manufacturers have conveyed these updated pricing details to key data center clients.
Major data center operators, including Microsoft, Google, and Oracle, have been informed of these price adjustments by server makers. These firms purchase significant volumes of AI infrastructure to support their cloud and data center operations. The rising costs associated with memory components have emerged as a key factor influencing the reported increase in server prices. AI deployment demands large amounts of high-bandwidth and server memory, alongside GPUs, CPUs, networking gear, and storage. During 2026, demand for AI applications has maintained a tight supply in several memory categories.
TrendForce predicts that contract prices for conventional DRAM will rise by 13% to 18% in the third quarter of 2026. The firm also forecasts NAND Flash contract prices to grow by 10% to 15% in the same period. The research firm noted that server DRAM remains in short supply as manufacturers prioritize capacity for AI-related workloads. Although some long-term supply agreements have limited the extent of price increases, overall memory costs for servers continue to be elevated.
Memory Cost Increases Persist Amid High Demand for AI Servers
In May, Nvidia announced that Vera Rubin had reached full production status with server manufacturers and supply-chain partners worldwide. The company indicated that products utilizing Rubin would be available in the second half of 2026. The Vera Rubin platform integrates the Vera CPU, Rubin GPU, NVLink 6, ConnectX-9, BlueField-4, and Spectrum-6 networking technologies. This platform succeeds Grace Blackwell as Nvidia’s latest rack-scale architecture optimized for large AI tasks. System manufacturers are now developing Rubin-based solutions for cloud providers, AI research laboratories, and hyperscale data centers.
Meanwhile, Grace Blackwell continues to be a crucial part of Nvidia’s current data center lineup. The GB200 NVL72 model connects 36 Grace CPUs with 72 Blackwell GPUs within a single liquid-cooled rack. Nvidia designed this system to function as a unified NVLink computing domain for extensive AI models. The reported price increases for servers vary depending on configurations, with memory capacity and chip generation playing significant roles in pricing. As a result, the 15% figure does not apply uniformly to every Nvidia server model.
Expansion of Vera Rubin Production Affects Nvidia’s Server Ecosystem
Manufacturers of memory modules have continued shifting their production focus toward server and high-performance products as AI demand soaks up available capacity. TrendForce highlighted that this shift has reduced supply for some PC and consumer DRAM products. While shipments of servers remained robust through 2026, buyers continued stockpiling data center memory. Market analysts expect server DRAM supply to stay tight into 2027, as demand growth exceeds new supply. These market conditions underpin the pricing environment influencing Nvidia’s recent price increases.
Nvidia enters this pricing phase with record-setting data center revenue. In its fiscal first quarter ending April 26, the company reported total revenue of $81.6 billion, with data center sales reaching an all-time high of $75.2 billion, up 92% year-over-year. In May, Nvidia projected second-quarter revenue of approximately $91 billion, plus or minus 2%, when releasing its results. The company is scheduled to announce its fiscal second-quarter earnings on August 26.