NVIDIA and Nebius Group have entered a significant strategic partnership aimed at building a next-generation hyperscale cloud specifically tailored for the “agentic AI” era. To solidify this collaboration, NVIDIA is investing $2 billion in Nebius, signaling a high level of confidence in the company’s engineering expertise and its ability to scale AI infrastructure globally.

The partnership is deeply integrated across the entire technology stack, moving beyond simple hardware supply to include collaborative AI factory design, production software, and fleet management. A primary goal is to meet the massive global demand for high-performance compute by deploying over 5 gigawatts of capacity by 2030. This includes the build-out of “AI factories” in the U.S. and the early adoption of NVIDIA’s most advanced upcoming architectures, such as the Rubin platform, Vera CPUs, and BlueField storage systems.

Beyond raw power, the two companies are focused on optimizing the software side of AI. They plan to develop a best-in-class inference stack to help developers and enterprises deploy AI models more efficiently. According to leadership from both firms, this collaboration represents a shift toward clouds that are built for AI from the ground up—rather than adapted from general-purpose computing—to support the rapidly growing needs of AI builders worldwide.