Microsoft Strikes $19.4 Billion Deal with Nebius to Boost AI Computing Power

Microsoft has secured a deal worth up to $19.4 billion with Nebius Group NV, giving the tech giant access to more than 100,000 of Nvidia’s latest GB300 chips. The move comes as demand for artificial intelligence infrastructure accelerates and data center capacity struggles to keep pace.

By tapping Nebius’ computing resources, Microsoft aims to offload some of its internal AI training to external partners, freeing its own servers for customer-facing services. The strategy highlights how the company is leaning on “neocloud” providers—smaller, specialized infrastructure firms—to scale rapidly in a fiercely competitive AI race.

The deal is part of a broader push. Microsoft has already committed more than $33 billion to partnerships with neocloud firms including CoreWeave, Nscale, Lambda, and Nebius. These companies lease AI-focused computing power, enabling Microsoft to expand capacity more quickly than if it relied solely on building out new data centers.

Industry watchers say this approach not only helps Microsoft manage the global shortage of AI data center capacity but also offers financial flexibility. Treating these partnerships as operational expenses, rather than heavy capital investments, gives the company more agility in managing costs while chasing growth in AI services.

Scott Guthrie, head of Microsoft’s cloud division, said, “We are aggressively securing our position in the AI space. “We’ve made the decision that we don’t want to be constrained in terms of capacity.”

Following the announcement, Nebius shares rose 5.5% in New York trading, while Microsoft’s stock showed little movement. Analysts noted that alongside these deals, Microsoft is still investing heavily in its own infrastructure, including a major new data center project in Wisconsin.

Summary:

Microsoft’s $19.4 billion agreement with Nebius underscores both the urgency and the scale of its AI ambitions. With demand for AI computing power soaring, the company is blending partnerships with neocloud providers and its own infrastructure expansion to stay ahead of rivals. The deal reflects a broader industry trend of balancing speed, cost, and capacity in the global race to dominate AI.

Frequently Asked Questions (FAQs)

Q1. What is the Microsoft–Nebius deal about?

Microsoft has signed a deal worth up to $19.4 billion with Nebius Group NV to gain access to more than 100,000 Nvidia GB300 chips. This partnership is aimed at boosting Microsoft’s AI computing power for internal projects and AI services.

Q2. Why is Microsoft partnering with neocloud providers like Nebius?

By working with neocloud providers, Microsoft can quickly scale AI infrastructure without waiting to build new data centers. It also allows the company to free its own servers for customer-facing AI services.

Q3. How much has Microsoft invested in neocloud partnerships overall?

Microsoft has committed over $33 billion to neocloud providers including CoreWeave, Nscale, Lambda, and Nebius to expand AI computing capacity rapidly.

Q4. What are the benefits of using neocloud services for Microsoft?

Neocloud partnerships provide operational flexibility, allowing Microsoft to treat costs as operating expenses instead of capital investments. This approach speeds up expansion and manages financial risk while meeting rising AI demand.

Q5. How does this deal help Microsoft in the AI race?

Access to additional computing power enables Microsoft to accelerate AI training and deployment, ensuring it stays competitive in the fast-growing AI market.

Q6. Are there any impacts on Microsoft’s stock or Nebius shares?

Following the announcement, Nebius shares rose 5.5% in New York trading, while Microsoft’s stock remained largely unchanged.

Q7. Is Microsoft still expanding its own data centers?

Yes. In addition to neocloud partnerships, Microsoft continues to invest in its own infrastructure, including a large-scale data center project in Wisconsin.

Q8. What does this deal indicate about the broader AI industry?

The partnership reflects a wider trend where tech companies balance speed, cost, and capacity by combining internal infrastructure with specialized cloud providers to meet soaring AI demand.