ROSELAND, N.J. and YORKTOWN HEIGHTS, N.Y.: One of the first deployments of NVIDIA GB200 NVL72 at supercomputing scaleSupercomputer will leverage IBM Storage Scale System to power AI research and developmentROSELAND, N.J. and YORKTOWN HEIGHTS, N.Y., Jan. 15, 2025 /PRNewswire/ — CoreWeave, the AI Hyperscaler™, today announced its plans to deliver one of the first NVIDIA GB200 Grace Blackwell Superchip-enabled AI supercomputers to IBM (NYSE: IBM), equipped with NVIDIA GB200 NVL72 systems, interconnected with NVIDIA Quantum-2 InfiniBand networking. IBM will use CoreWeave’s highly performant, reliable, and resilient cloud platform to train the next generations of its Granite models, IBM’s open source, enterprise-ready series of AI models that deliver state-of-the-art performance relative to model size while maximizing safety, speed, and cost-efficiency for enterprise use cases.