Oracle Cloud Infrastructure (OCI) has deployed thousands of NVIDIA Blackwell GPUs in its data centers, marking a significant step in the advancement of AI and reasoning models. The first wave of these powerful GPUs is now optimized and ready for customer use on NVIDIA DGX Cloud and OCI. The deployment involves state-of-the-art NVIDIA GB200 NVL72 racks, each of which integrates 36 NVIDIA Grace CPUs and 72 NVIDIA Blackwell GPUs. This setup ensures scalable, low-latency performance thanks to high-speed NVIDIA Quantum-2 InfiniBand and Spectrum-X Ethernet networking. Additionally, the systems are supported by a comprehensive stack of software and database integrations from both NVIDIA and OCI. OCI, known as one of the largest and fastest-growing cloud service providers globally, is among the first to embrace this cutting-edge technology. The company aims to construct one of the world’s largest Blackwell clusters, which will eventually exceed 100,000 GPUs. This ambitious expansion is driven by the rapidly increasing demand for inference tokens and accelerated computing, reflecting the ongoing surge in AI innovation. Companies like OpenAI have recently introduced new reasoning models, further intensifying this demand. NVIDIA Grace Blackwell systems are transforming cloud data centers into efficient AI factories capable of generating intelligence on a massive scale. The GB200 NVL72 platform, which combines powerful CPU and GPU capabilities, is designed to deliver exceptional performance and energy efficiency, making it ideal for running agentic AI applications. These systems not only support the development of AI but also handle complex tasks such as training reasoning models, autonomous vehicle development, and accelerating chip design and manufacturing. OCI provides flexible deployment options to cater to a wide range of customers, including major technology firms, enterprise users, government agencies and contractors, and regional cloud providers. Customers can access Blackwell GPUs in public, government, and sovereign clouds, or even in their own data centers through OCI Dedicated Region and OCI Alloy, at any scale they require. Several notable customers are eager to begin using the new GB200 systems. These include large technology companies, enterprise clients, and various governmental organizations. The versatility of the deployment options ensures that the technology can be tailored to meet diverse needs, from small-scale projects to large, resource-intensive applications. NVIDIA DGX Cloud, the platform that offers optimized software, services, and technical support for AI development and deployment, is the source of these racks. The immediate availability of the GB200 NVL72 racks from DGX Cloud and OCI signifies the readiness of these systems to tackle a variety of AI projects, from model training to tool development. The integration of NVIDIA’s GB200 NVL72 racks into Oracle’s infrastructure highlights the collaborative effort between leading tech companies to push the boundaries of AI. This partnership is crucial in meeting the growing computational demands and fostering further innovation in the field. With the deployment now complete, the stage is set for a new era of AI-powered advancements across multiple industries.