Nvidia trims HBM4 bandwidth for Rubin platform, sources say
Nvidia is reportedly adjusting the fourth-generation high-bandwidth memory (HBM4) specification for its Vera Rubin NVL72 platform, cutting bandwidth per GPU from the previously announced 22TB/s to 19.2TB/s, a reduction of about 13%. South Korean industry observers said the move could increase supply flexibility and help balance performance with mass-production delivery, although the reason for the change and its actual impact on supplier orders remain unconfirmed.
Read original article ↗
Related Articles
Xyte Reports Record Quarter as Enterprise Demand Grows
Best-ever September quarter driven by enterprise demand and expanding dealer adoption as Xyte names Andrew Gross CEO. HO
Ananti Jeju opens new clubhouse blending architecture with 'gotjawal' forest
Ananti has opened a new clubhouse at Ananti Jeju, integrating "gotjawal," Jeju Island’s distinctive forest ecosystem sha
NCS opens AI Mission Centre in Ang Mo Kio, to triple embedded AI engineers to over 300
The Singtel-owned firm also announced three partnerships to scale up AI projects at Changi Airport, the Institute of Men