CoreWeave Brings Nvidia's Vera Rubin NVL72 to its Cloud Services
CoreWeave has announced that Nvidia's Vera Rubin NVL72 rack-scale AI platform is now available in its cloud services. The move is part of a series of announcements made by the company at its Fully Connected conference in San Francisco.
Cognition, the developer of Devin AI software-engineering agent, is the first customer to use CoreWeave's Vera Rubin cluster. According to Cognition, running on Vera Rubin NVL72 resulted in a 4.8X increase in total token throughput for SWE-2 inference workloads compared to an Nvidia GB200 NVL72 baseline.
CoreWeave has also added support for the Vera CPU, which is designed specifically for AI agents and modern AI systems. The company's initial Vera deployment features rack-scale configurations containing 128 Vera CPUs, or 11,264 CPU cores in a single rack.