CoreWeave Becomes First to Deploy Nvidia's Vera Rubin NVL72 System
CoreWeave has become the first AI cloud provider to deploy Nvidia's Vera Rubin NVL72 system, connecting hundreds of next-generation GPUs into a production cluster designed for demanding AI workloads. This achievement has sent CoreWeave's stock (CRWV) surging 14% with trading volume nearly double its three-month average.
The company received the NVL72 hardware on May 31 and had it operational by June 1, a turnaround time of just 24 hours. Each NVL72 rack packs 72 Rubin GPUs and 36 Vera CPUs, all connected through NVLink 6 interconnect running at 260 TB/s of bandwidth.
The system delivers up to 3.6 exaFLOPS of NVFP4 inference capability per rack and offers a 10x improvement in inference throughput per megawatt compared to the previous Blackwell generation. CoreWeave claims that clusters can expand to thousands of GPUs using either Spectrum-X Ethernet or Quantum-X800 InfiniBand networking.