- Initial contract of $17.400 billion, expandable to $19.400 billion, for AI infrastructure with GPUs.
- Service from the new data center in Vineland, New Jersey, with phased deployment between 2025 and 2026 and a horizon through 2031.
- Nebius will finance part of the project with cash flow from the contract and short-term debt secured by the agreement.
- Strong stock market reaction for Nebius and knock-on effect for partners in the "neoclouds" ecosystem.
The AI computing ecosystem adds a significant new piece with the agreement between two key players. The Redmond-based tech giant will rely on dedicated Nebius capacity to bolster its AI offering in the cloud, a step that reinforces its strategy of combining its own resources with specialized partners. In figures, this is a major commitment and with a multi-year roadmap.
According to documents and communications from both companies, a contract has been signed five years whose value starts at 17.400 million, with the option to expand capacity in another 2.000 million If demand requires it. The service will be provided using GPU-equipped infrastructure from Nebius facilities in the United States, with a phased rollout schedule that extends operational milestones. until 2031.
A major agreement on the AI cloud
The agreement provides for Nebius to supply Microsoft with accelerated computing capacity Next-generation GPUs from its new data center Vineland, New JerseyThe provision will be organized in multiple batches, with first phases between 2025 and 2026, and room for expansion based on workload needs.
Beyond the economic volume, the alliance is interpreted as a measure to ensure availability at a time when demand for model training and inference regularly exceeds supply. With this movement, Microsoft secures additional capacity to support Azure AI and its own projects, in addition to commitments with partners in the AI ecosystem.
Explicitly, the European company confirms that the contract can be scaled up to 19.400 million if additional capacity options are activated. This layered design allows supply to be adjusted to actual demand, a practice that is becoming increasingly common in supply chain agreements. specialized infrastructure.
Why Microsoft is turning to "neoclouds"
The bet fits with a clear trend: instead of building all the computing muscle internally, Microsoft combines its data center network with "neocloud" providers that can deploy AI capabilities quickly.This approach complements existing investments and reduces bottlenecks when demand surges.
The move also sends a signal to the market: there is room for specialized players that offer dedicated capacity with GPUs, high speed networks and clusters optimized for large models. Nebius' example adds to other agreements that have acted as an expansion valve for AI workloads, and reinforces the idea of a computing supply chain more diversified.
Impact on Nebius: funding, deployment, and timelines
The Amsterdam-based company details that part of the investment effort will be covered with cash flow generated by the contract itself, supplemented with short-term debt supported by the agreement. This scheme allows for accelerated launch of the Vineland campus without straining liquidity.
The rollout will be phased, with technically defined milestones to gradually turn on capacity as the accelerated hardware batches arrive and the interconnections are completed. The roadmap establishes the main ramp between 2025 and 2026, while the contractual horizon extends operational commitments until 2031.
In the words of the management team—paraphrased—the economics of the agreement are attractive in themselves and, in addition, should boost growth of the AI cloud business from 2026Management anticipates more large contracts with laboratories and large technology companies if current demand continues.
- Technical validation: Microsoft's entry as a supplier strengthens Nebius' credibility in AI workloads.
- Funded capacity: The contract helps fund the expansion of the new data center and the acquisition of GPUs.
- International scale: The agreement positions Nebius to compete in large computing tenders.
Market reaction and implications for the sector
The market response was immediate: Nebius shares advanced around 35% during the session and they surpassed the 40% in extended operations, with specific levels close to $92,49 after the close. The movement dragged other AI computing providers higher, with increases of around 5% in comparable companies.
Analysts interpret the pact as A catalyst for the neocloud segment, which is gaining visibility and capital flows. Even so, they warn that Valuations could heat up quickly if the expected operating margin does not materialize. as the committed capacity is activated.
- Opportunity: direct exposure to critical infrastructure for generative AI and deep learning.
- Risk: execution and scaling costs; profitability will depend on the pace of deployment and utilization.
Where Nebius and its European focus come from
Nebius emerged after the restructuring of international assets linked to Yandex and its subsequent separation from the Russian business. CEO Arkady Volozh reorganized the perimeter, brought in a large group of engineers, and focused the plan on cloud infrastructure for AI. The company resumed trading on the Nasdaq in 2024 under a new ticker and has since gained traction among AI clients.
The firm has placed its emphasis on Europe and North America, with the development of data centers and projects that include clusters of NVIDIA GPUs and high-speed networks. This new contract with Microsoft serves as a lever to expand its footprint, serve strategic customers—including key European players—and consolidate its role in the supply chain. advanced computing.
The understanding between Nebius and Microsoft marks a turning point: it guarantees additional capacity for Redmond's AI roadmap and, at the same time, gives Nebius the volume, visibility and funding to accelerate its expansion. If the deployment milestones are met and the demand for models continues to grow, the pact could become a benchmark for how the AI is articulated new AI infrastructure between hyperscale and specialized providers.
I am a technology enthusiast who has turned his "geek" interests into a profession. I have spent more than 10 years of my life using cutting-edge technology and tinkering with all kinds of programs out of pure curiosity. Now I have specialized in computer technology and video games. This is because for more than 5 years I have been writing for various websites on technology and video games, creating articles that seek to give you the information you need in a language that is understandable to everyone.
If you have any questions, my knowledge ranges from everything related to the Windows operating system as well as Android for mobile phones. And my commitment is to you, I am always willing to spend a few minutes and help you resolve any questions you may have in this internet world.