The AI hype has taken HBM demand to new heights and is expected to double by next year along with significant price hikes of up to 10%.
Hunger For AI Computing Power Has Led HBM Supply To Grow Rapidly, Unit Prices Soar To New Levels
Research outlet, TrendForce reports that the memory industry has immensely benefited from the demand for AI-focused activities, with HBM sales playing a vital role in reviving the markets. Avril Wu, TrendForce's Senior Research Vice President, has disclosed that HBM has witnessed a colossal soar in revenue share in the memory industry, outperforming both consumer and client DRAM sales. The unit price of HBM is said to be significantly higher than what DRAM is going for, which shows how dominating the standard has been in recent times.
Moving into statistics, the share of HBM in the DRAM markets is expected to outgrow going into 2025, and it may reach double digits, because of how rapidly the markets have adopted the standard, primarily due to the advent of the AI hype, which has brought in the need for significant compute power, ultimately catalyzing4 the sales of AI accelerators, where HBM is utilized as a primary component.
As we proceed to the next-gen era of AI computing, firms like AMD and NVIDIA have shown tremendous interest in newer HBM processes, such as the popular HBM3, which has already seen integration in AMD's Instinct MI300X AI accelerators, & from NVIDIA's camp, the likes of the Hopper H200 and Blackwell B100 AI GPUs infused with HBM3E memory. Memory manufacturers are also prepping to launch the next-gen HBM4 products by 2025-2026.
This has created tremendous demand and raised HBM's per-unit prices. However, the inefficiencies involved with HBM manufacturing have held back high-volume production, and the rising costs have been unfavorable for buyers.
Despite that, HBM demand is expected to grow by a whopping 200% per year, showing how big the memory standard will be, given that the AI hype continues to maintain its trend.
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
