SANTA CLARA, CALIFORNIA / RankWire.AI / – Nvidia is set to implement price hikes exceeding 15% on numerous AI server configurations scheduled for early 2027. These price adjustments apply to systems built on the Vera Rubin and Grace Blackwell platforms. The extent of each increase depends on the specific chip generation and memory configuration. Nvidia has not officially announced a blanket price rise for all server products but has shared the new pricing details with major data center clients.
Major data center operators, including Microsoft, Google, and Oracle, have been informed of these pricing changes. These companies purchase large quantities of AI infrastructure to support their cloud and data center services. The rising costs of memory have emerged as a key factor influencing the reported increases in server prices. AI deployment demands significant amounts of high-bandwidth and server memory, along with GPUs, CPUs, networking hardware, and storage. Throughout 2026, the demand from AI projects has kept many memory categories in tight supply.
Forecasts from TrendForce indicate that conventional DRAM contract prices are expected to grow between 13% and 18% in the third quarter of 2026. Additionally, NAND Flash contract prices are projected to see a 10% to 15% increase during the same period. The research firm pointed out that server DRAM remains undersupplied as manufacturers prioritize capacity for AI applications. Although some long-term supply agreements have restrained price hikes, overall server memory costs continue to be high.
Memory Cost Growth Driven by Sustained High Demand for AI Servers
In May, Nvidia announced that Vera Rubin had entered full production with server manufacturers and supply-chain partners globally. The company indicated that products utilizing 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 tech. This platform succeeds Grace Blackwell as Nvidia’s latest rack-scale architecture designed for intensive AI workloads. System developers are creating Rubin-based products for cloud providers, AI laboratories, and hyperscale data centers.
Meanwhile, Grace Blackwell remains a key component of Nvidia’s current data center portfolio. The GB200 NVL72 design pairs 36 Grace CPUs with 72 Blackwell GPUs within a single liquid-cooled rack. This setup is engineered to operate as a unified NVLink computing domain for large AI models. The reported increases in server prices vary depending on configuration, with memory capacity and chip generation being critical factors. Consequently, the 15% figure does not represent a uniform increase across all Nvidia server offerings.
Expansion of Vera Rubin Production Boosts Nvidia’s Server Ecosystem
Suppliers of memory continue to shift production focus toward server and high-performance components, driven by the high demand from AI applications. According to TrendForce, this shift has reduced supply for some PC and consumer DRAM products. While server shipments have remained robust throughout 2026 and buyers have been expanding their data center memory inventories, the market researcher anticipates that server DRAM supply will stay tight into 2027. Growing demand outpaces new supply, creating the cost environment that influences Nvidia’s reported price adjustments.
Nvidia begins this pricing period with record-breaking data center sales. For its fiscal first quarter ending April 26, the company reported a total revenue of $81.6 billion, with data center revenue reaching an all-time high of $75.2 billion, representing a 92% increase from the previous year. In May, Nvidia projected second-quarter revenue at around $91 billion, plus or minus 2%, when announcing these results. The company is scheduled to release its fiscal second-quarter earnings on Aug. 26.