【AI+NVDA】NVIDIA Reinvests $2 Billion in "Circular Investment" - Nebius Deploys 5 GW GPU

NVIDIA (US: NVDA) has once again reached a “cycle investment” agreement, investing $2 billion in cloud computing platform Nebius (US: NBIS). In return, Nebius plans to deploy 5-gigawatt GPU power by the end of 2030. The two parties will jointly develop the next-generation large-scale cloud computing platform.

Nebius’s stock rose 16%, reaching $111.75; NVIDIA increased by 0.8%, currently at $186.13.

Recently, NVIDIA has invested in several companies. Last week, it reached agreements with two U.S. optical technology companies, Lumentum (US: LITE) and Coherent (COHR), each investing $2 billion to support the development and production of advanced optical technologies. In December of last year, NVIDIA invested $2 billion in chip design company Synopsys (US: SNPS).

As part of the deal, both sides will collaborate on AI infrastructure deployment, cluster management, model inference, and AI factory design and support. This covers the complete technology chain from the underlying “AI factory architecture” to “production software,” aiming to accelerate the creation of an industry-leading full-stack AI cloud platform. Multiple gigawatt-level computing centers will also be deployed across the U.S.

According to the cooperation agreement, the two parties will focus on four core areas:

• AI Factory Design and Support: Including access to partner design materials, participation in design review processes and acceptance, early sample and system software support, launch support, and regular business and technical reviews with system partners.
• Inference Stack: Combining NVIDIA’s latest software technologies, optimized models, and libraries to create a top-tier inference and agent-based AI stack for developers and enterprises.
• AI Infrastructure Deployment: Deploying multiple generations of NVIDIA infrastructure on the Nebius platform, including NVIDIA Rubin platform, Vera CPUs, and NVIDIA BlueField storage systems, through early adoption of NVIDIA’s computing architecture.
• Cluster Management: Deploying NVIDIA’s latest GPU health monitoring software and recommendations to optimize the overall health of Nebius’s clusters.