NVIDIA Vera Rubin Driving Performance Per Watt, Lowest Token Cost for Partners Worldwide
NVIDIA has begun production of its Vera Rubin NVL72 rack-scale systems, which are currently being deployed across data centers operated by CoreWeave, Google Cloud, Microsoft Azure, and Oracle Cloud Infrastructure. By focusing on performance per watt and reduced token costs, this hardware rollout aims to improve the efficiency and economic scale for global partners training and deploying large-scale artificial intelligence models.
Covered by 1 source
- NNVIDIA AI Blog↗NVIDIA Writers9h ago