Nvidia is putting another major stake behind CoreWeave, investing $2 billion as the data center company races to expand AI computing capacity. The move gives CoreWeave fresh support at a time when its rapid growth, heavy borrowing and central role in AI infrastructure are all under close attention.
A larger bet on AI compute
Nvidia said on Monday it invested $2 billion in CoreWeave to speed up the company’s effort to add more than 5 gigawatts of AI computing capacity by 2030. Nvidia was already an investor in CoreWeave, but this deal pushes the relationship further into buildout, hardware and software integration.
The chipmaker bought CoreWeave’s Class A shares at $87.20 per share. As part of the arrangement, CoreWeave and Nvidia plan to build “AI factories,” the data centers that will use Nvidia products to serve AI workloads.
For CoreWeave, the investment is more than a financial headline. It links one of the most visible AI infrastructure companies with the chipmaker whose products sit at the center of the current AI buildout. It also signals that Nvidia wants CoreWeave to remain a major channel for deploying its newest systems into cloud businesses and enterprises.
What CoreWeave will add to its platform
The deal includes a broad product integration plan. CoreWeave will bring Nvidia products across its platform, including the new Rubin chip architecture, which is set to replace the current Blackwell architecture.
The company will also integrate Bluefield storage systems and Nvidia’s new CPU line, Vera. In practical terms, the partnership is not limited to GPUs. It reaches into compute architecture, storage systems and CPU infrastructure.
Nvidia will also help CoreWeave buy land and power for data centers. That part of the arrangement matters because AI data centers are not only a chip supply problem. They also require physical sites, access to energy and coordinated planning before any compute capacity can come online.
The companies will also work together to include CoreWeave’s AI software and architecture within Nvidia’s reference architecture. The target customers are cloud businesses and enterprises, the same market that needs large-scale infrastructure for AI training and inference.
Why the deal arrives under scrutiny
The investment comes after months of scrutiny over CoreWeave’s debt-funded expansion. The company has raised billions in debt as it continues building out its data center operations.
As of September 2025, CoreWeave had $18.81 billion in debt obligations, according to PitchBook data. The company reported revenue of $1.36 billion in the third quarter.
CoreWeave CEO Michael Intrator has defended the company’s model, which funds operations by raising debt with its GPUs as collateral. He has also responded to concerns about circular deals in the AI industry by saying companies have to “work together” to address a “violent change in supply and demand.”
That context is central to understanding why Nvidia’s backing matters. CoreWeave is expanding into a market with strong demand for AI training and inference, but it is doing so through a capital-heavy model. The new investment does not remove those questions, but it does show that Nvidia is willing to deepen its support while CoreWeave continues to scale.
From crypto mining to AI infrastructure
CoreWeave has ridden the AI wave after transitioning from a crypto mining company into a provider of data center services for AI training and inference. Since its IPO in March last year, the company has been building out more of its technology stack through acquisitions.
It acquired Weights & Biases, an AI developer platform, in March. Soon after, it bought reinforcement learning startup OpenPipe. In October, it agreed to acquire Marimo, an open source Jupyter notebook competitor, and Monolith, another AI company.
CoreWeave also recently expanded its cloud partnership with OpenAI. The company currently counts several hyperscalers as customers, including OpenAI, Meta, and Microsoft.
Together, those moves show a company trying to offer more than raw data center capacity. CoreWeave is adding developer tools, AI software pieces and cloud relationships around the infrastructure it is building.
A signal for the wider AI market
CoreWeave’s shares were up more than 15% following news of the deal. That reaction reflects how closely investors are watching AI infrastructure spending and the companies supplying it.
For Nvidia, the CoreWeave investment is part of a broader pattern. The chipmaker has made several dozen investments in the past year as it continues fueling investment in and development of AI technology.
The deal also shows how tightly connected the AI infrastructure market has become. Nvidia supplies the hardware and architecture. CoreWeave builds and operates the data center services. Cloud businesses and enterprises are the customers the companies want to reach.
The immediate takeaway is straightforward: Nvidia is using capital, products and planning support to help CoreWeave expand faster. The larger question is how sustainable that expansion remains as AI demand, data center capacity and debt-funded growth continue to move together.