Nvidia AI Server Prices to Surge Over 15% in 2024
By Business Desk
Nvidia AI server prices are set to increase by over 15% in early 2024 due to rising memory chip costs, impacting systems with Vera Rubin and Grace Blackwell chips.
Nvidia Corp. customers are facing a significant price increase for AI servers, with costs projected to rise by over 15% starting early next year. This hike is primarily driven by soaring memory chip expenses, affecting systems equipped with the company’s advanced Vera Rubin and Grace Blackwell chips.
- Price Increase: Over 15%
- Affected Systems: Vera Rubin and Grace Blackwell chips
- Nvidia Gross Margin: 75%
The exact percentage of this increase will vary based on the Nvidia chip generation and memory configurations. Server builders, who supply major data center operators including Microsoft Corp., Alphabet Inc.’s Google, and Oracle Corp., have already begun informing their clients of these upcoming adjustments. Nvidia has not yet issued public comments on the situation.
Memory Chip Leverage
This development underscores the substantial market power held by key memory chip manufacturers amidst an unprecedented surge in demand for AI infrastructure.
- Samsung Electronics Co.
- SK Hynix Inc.
- Micron Technology Inc.
Despite efforts to ramp up production, these companies struggle to meet the escalating demand for dynamic random access memory (DRAM), leading to significant price increases. Nvidia’s accelerator processors are essential for computers developing and operating AI software, with their efficiency directly tied to the paired DRAM quantity.
The company maintains a highly profitable position within the semiconductor sector. Nvidia’s contract manufacturer, Taiwan Semiconductor Manufacturing Co., cannot fully satisfy the persistent, runaway demand, allowing Nvidia to charge tens of thousands of dollars per chip.
Market Dynamics and Customer Response
The current market lacks viable alternatives to Nvidia’s offerings, partly fueling the significant rise in AI accelerator prices. Nvidia has also recently implemented price increases for its gaming-oriented PC graphics cards.
The industry will closely monitor how Nvidia’s customers respond to these higher costs, particularly regarding potential opportunities for competitors. This hinges on their ability to independently secure adequate memory chips. Major clients like Amazon, Microsoft, Google, and Meta are developing their own in-house chip programs, yet still rely on Nvidia for data center expansions.
Progress towards greater independence for these tech giants will depend on their access to supply from Samsung, SK Hynix, and Micron. These price increases are expected to add further complexity to ambitious AI data center build-outs already facing multiple challenges. Existing industry hurdles include project delays, labor shortages, tightening capital markets, and community opposition. Investors and the technology sector are keenly anticipating Nvidia’s fiscal second-quarter earnings release next week.