2026年夏季，英伟达“Vera Rubin”AI系统将正式出货，采用VR200 NVL72服务器方案，用于驱动下一代大模型。该系统HBM4内存由SK海力士（占70%）和三星（30%）供应，美光未进入HBM4供应链，因其未通过英伟达带宽规格认证——系统最终带宽达22 TB/s。美光转而为系统内“Vera”CPU提供LPDDR5X内存，最高达1.5TB，以弥补HBM4缺位影响。
Source Link2026年夏季，英伟达“Vera Rubin”AI系统将正式出货，采用VR200 NVL72服务器方案，用于驱动下一代大模型。该系统HBM4内存由SK海力士（占70%）和三星（30%）供应，美光未进入HBM4供应链，因其未通过英伟达带宽规格认证——系统最终带宽达22 TB/s。美光转而为系统内“Vera”CPU提供LPDDR5X内存，最高达1.5TB，以弥补HBM4缺位影响。
Source LinkDisclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.