Nvidia's Vera Rubin AI Platform Reaches Full Production with Fall Shipments

Nvidia has announced that its next-generation Vera Rubin AI computing platform is ramping up to full production, with the first customer shipments set to begin this fall. The announcement marks a significant milestone for the company's most ambitious data-center platform to date, which will be used in cloud data centers.

The Rubin platform is built around two custom parts, a Vera CPU and a Rubin GPU, paired inside a rack-scale system called NVL72. Each Rubin GPU carries 288GB of HBM4 memory with roughly 22 terabytes per second of memory bandwidth per GPU, connected through NVLink 6 at 3.6 terabytes per second per GPU.

Nvidia has named a long list of partners for the platform, including Dell Technologies, HPE, Lenovo, and Supermicro as lead builders, alongside other manufacturers and storage partners such as NetApp, VAST Data, and WEKA. On the cloud side, Nvidia names CoreWeave, Microsoft Azure, Lambda, Nebius, Nscale, IBM Cloud, GMI Cloud, IREN, Firmus, SpaceXAI, and Vultr as adopters of the platform's confidential-computing features specifically.