SANTA CLARA, CALIFORNIA / RankWire.AI / – Nvidia is planning to implement price hikes exceeding 15% on numerous AI server configurations scheduled for delivery in early 2027. These adjustments impact systems based on Vera Rubin and Grace Blackwell technology. The final increase percentages vary depending on factors such as chip generation, memory capacity, and overall system design. Nvidia has not announced a uniform, companywide price increase applicable to all server models. Instead, hardware manufacturers responsible for assembling AI systems have relayed updated pricing information to large data center clients.
Microsoft, Google, and Oracle are among the leading cloud service providers purchasing substantial quantities of accelerated computing hardware. Their data centers utilize AI servers for tasks like model training, inference, and cloud-based services. Throughout 2026, memory has emerged as one of the most significant cost pressures across these systems. Modern AI servers integrate GPUs with high-bandwidth memory, server DRAM, storage, and rapid networking. The high demand for these components has kept supply shortages persistent across various segments of the memory market.
TrendForce predicts that contract prices for conventional DRAM will grow by 13% to 18% during the third quarter of 2026. It also forecasts NAND Flash contract prices to increase between 10% and 15% over the same timeframe. Server DRAM, in particular, remains constrained as memory manufacturers allocate more manufacturing capacity to AI and data center applications. The rising memory costs have driven up the expense of constructing advanced computing systems, contributing to the overall pricing environment for next-generation AI servers.
Memory price hikes exert additional pressure on AI infrastructure costs
According to Nvidia, Vera Rubin entered full production with server manufacturers and supply-chain partners in 2026. Products utilizing this platform are expected to become available during the second half of the year. Rubin combines the Vera CPU with the Rubin GPU, NVLink 6, and various networking technologies, targeting large-scale AI workloads in cloud and hyperscale data centers. It follows Grace Blackwell as Nvidia’s latest rack-scale computing platform.
Grace Blackwell remains integral to existing AI data center deployments. The GB200 NVL72 system links 36 Grace CPUs with 72 Blackwell GPUs within a liquid-cooled rack. Nvidia designed this platform to function as a unified NVLink computing domain. Price variations for these systems depend on specific hardware configurations rather than a fixed percentage. Factors like memory capacity, processor generation, and rack design all influence the final cost of each individual server setup.
Demand for servers sustains tight memory supply conditions
Manufacturers of memory components have shifted a larger share of production toward server and high-performance products as artificial intelligence demands grow. TrendForce reports that this shift has reduced the supply available for some PC and consumer memory markets. Data center operators continued purchasing substantial quantities of server memory throughout 2026. The firm anticipates that server DRAM availability will remain limited into 2027 as demand outpaces new supply. This situation continues to influence component prices across AI infrastructure.
Nvidia is entering this pricing adjustment period following another quarter of record revenue from data centers. The company announced fiscal first-quarter revenue of $81.6 billion for the period ending April 26, 2026. Revenue from Data Center operations reached $75.2 billion, representing a 92% increase compared to the same quarter last year. Nvidia has also projected second-quarter revenue to be around $91 billion, plus or minus 2%. Its fiscal second-quarter results are scheduled for release on Aug. 26, providing the latest insights into its financial performance.