Skip to main content

Microsoft Partners with Nebius in $17.4B AI Infrastructure Mega-Deal

·503 words·3 mins
Microsoft Nebius AI Infrastructure Cloud Computing GPUs
Table of Contents

Microsoft and Nebius Group have announced a massive AI infrastructure agreement worth up to $17.4 billion, with an option for Microsoft to purchase an additional $2 billion in services through 2031.

Following the news, Nebius’s stock price surged by over 60% in after-hours trading, eventually settling at a gain of 43%-50%, pushing its market capitalization past $15 billion.

Under the agreement, Nebius will deploy dedicated computing capacity for Microsoft at its new data center in Vineland, New Jersey. These GPU services will be rolled out in several phases throughout 2025 and 2026, with the earliest services available in late 2025. Each batch must meet strict deployment and availability standards. Microsoft has not disclosed exactly how it will use Nebius’s resources.

Microsoft’s Computing Crunch
#

Behind this massive investment lies Microsoft’s severe AI computing shortage.

In July, Microsoft’s CFO Amy Hood warned investors that the company expects to face capacity constraints through the end of 2025.

This shortage has impacted the rollout of Microsoft’s own AI products and limited its ability to deliver AI services via Azure Cloud. Despite billions spent on new infrastructure, demand continues to outpace supply.

To solve this, Microsoft has adopted a multi-vendor strategy. Alongside Nebius, Microsoft has signed AI computing contracts with OpenAI, CoreWeave, and other providers to secure capacity.

Nebius’s Technical Background
#

Nebius isn’t an unknown player—it’s a spin-off of Russian internet giant Yandex. Founded in 1989 and now headquartered in Amsterdam, the company rebranded from Yandex NV to Nebius last year.

Key investors include NVIDIA and Accel Partners, providing strong financial and technical backing. Nebius focuses on AI cloud services powered by NVIDIA GPUs, offering integrated hardware-software solutions.

In November, Nebius opened offices in San Francisco, Dallas, and New York. In a blog post, the company wrote:

“Expanding our presence in the U.S. brings us closer to customers and allows us to play a role in supporting the growth of innovative American AI companies.”

Reshaping the AI Market Landscape
#

For Microsoft, the deal helps directly address its AI computing shortage. By locking in capacity, the company ensures it can meet the growing demand for Azure AI services.

This move highlights a shift in the AI infrastructure market: tech giants are building deep partnerships with specialized vendors rather than relying solely on traditional cloud providers. Microsoft’s dual strategy—being the largest customer of both CoreWeave and now Nebius—shows its intent to diversify supply chains and avoid over-reliance on a single partner.

For Nebius, the deal is transformative. Founder and CEO Arkady Volozh said:

“The economic benefits of this deal are attractive on their own, but more importantly, it will help us accelerate our AI cloud business in 2026 and beyond.”

The company plans to fund expansion with cash from the deal, contract-backed debt, and possibly additional financing—aiming to grow faster than originally planned.

Key Takeaway:
Microsoft’s $17.4B deal with Nebius isn’t just about GPUs—it’s about reshaping the global AI infrastructure race. As computing shortages intensify, partnerships like this will define the future of cloud AI services.

Related

NVIDIA’s $1.5B Strategy: Renting Its Own GPUs from Lambda
·472 words·3 mins
NVIDIA AI GPUs Lambda CoreWeave Cloud Computing
Cisco, NVIDIA, and VAST Launch Enterprise-Grade Agentic AI Factory
·623 words·3 mins
CISCO NVIDIA Vast Data Agentic AI AI Infrastructure RAG Acceleration
AMD Financial Analyst Day 2025: Zen 6, RDNA 5, and AI Roadmap
·605 words·3 mins
AMD Zen 6 Zen 7 RDNA 5 Instinct MI600 AI Financial Analyst Day 2025 GPUs CPUs