Nvidia's Rubin expected to use mostly 12-layer HBM4 despite reported 8-layer shift

Nvidia's next-generation Rubin AI accelerators are expected to rely predominantly on 12-layer HBM4 high-bandwidth memory, despite reports that the chipmaker has asked Samsung Electronics and SK hynix to increase supplies of 8-layer products.

Covered by 2 publishers within 9 hours of the first report.

Summary from DigiTimes.

Reporting2

DT DigiTimes Nvidia's Rubin expected to use mostly 12-layer HBM4 despite reported 8-layer shift · DIGITIMES 9:32 PM

WCF Wccftech NVIDIA Rubin GPUs Expected To Stick With 12-Hi HBM4 Despite Reports Of An 8-Hi Shift, As HBM Prices Could Surge 121% In 2027 · Sebastian Castellanos 6:50 AM

Related

SA Nvidia, Samsung back $90M round for AI agent startup Nous Research October 7

UAI Mecka Raises $60M Series B to Scale Robot Data and Deployment October 7

VG Volvo will use Nvidia’s tech for hands-free, point-to-point driving October 15

HPC Bull to Bring NVIDIA Vera and Vera Rubin Ultra NVL72 to Europe October 15

Topics SamsungNvidia