•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Nvidia’s role in AI infrastructure is well known: it supplies the GPUs, hyperscalers build data centers, and software companies develop applications. Less visible is the middle layer—specialized cloud providers that rent GPU compute to AI model builders. Among that group, Nebius Group stands out for the size of its contracted backlog relative to its current scale.
Nebius Group is not a typical startup. It emerged in 2024 from the restructuring of Yandex, the Russian internet company. Its founding CEO, Arkady Volozh, brought extensive infrastructure experience, including decades of operating large-scale data center systems, and entered the venture with $2.5 billion in capital and hundreds of infrastructure engineers.
The company is vertically integrated compared with many cloud competitors. Nebius designs its own proprietary server racks, builds InfiniBand-based networking software called Nebius Fabric, and operates data centers in the United States and Europe. The stated goal is to deliver high-performance AI compute with lower latency and more competitive pricing than general-purpose cloud providers.
The contract timeline is central to Nebius’ growth outlook. In September 2025, Nebius signed a five-year contract with Microsoft worth up to $19.4 billion. In December 2025, it added a $3 billion five-year contract with Meta Platforms. In March 2026, Meta expanded that deal to up to $27 billion.
According to the article, the expanded contract covers $12 billion in dedicated capacity and up to $15 billion in additional available capacity, and it represents one of the first large-scale deployments of Nvidia’s Vera Rubin platform. In the same week, Nvidia announced a $2 billion direct equity investment in Nebius as a strategic partner for next-generation hyperscale AI infrastructure.
When aggregated, Nebius is described as having a total contracted backlog approaching $50 billion for the 2027–2031 period, compared with 2025 revenue of $530 million.
At trading levels around $164, Nebius Group carries a market capitalization of approximately $41 billion. Analysts’ price targets range from $143 to $211, with 27 buy ratings and essentially no sell ratings.
The bullish case presented in the article links revenue conversion and capacity scaling. It cites scaling data center capacity from 170 megawatts at the end of 2025 toward a target of 800 megawatts to 1 gigawatt by year-end 2026. Management guidance for 2026 revenue is $3 billion to $3.4 billion, with adjusted EBITDA margin near 40%.
A “10x” outcome from the current market capitalization would imply a valuation of roughly $390 billion, which the article characterizes as requiring Nebius to become a dominant AI-focused cloud platform—analogous to Amazon Web Services for general cloud computing, but specialized for AI. The article notes that such an outcome is not guaranteed.
The article highlights multiple risks that could affect whether the contracted backlog translates into results. Nebius plans to spend $16 billion to $20 billion in capital expenditures this year, and the spending is described as moving faster than current revenue.
It also points to execution dependencies: the infrastructure must be built on time; Microsoft and Meta must not renegotiate or reduce capacity usage; and Nvidia’s latest GPU platforms—including the Vera Rubin NVL72, which Nebius is set to deploy in H2 2026—must perform as expected in production environments.
The article emphasizes that these are not minor uncertainties.
Despite the risks, the article argues that Nebius combines validated technology—reinforced by Nvidia’s $2 billion investment—with a large contracted revenue foundation and a founding team with deep infrastructure experience. It frames Nebius as a rare mix: large enough to be taken seriously, small enough that growth could still be transformative, and differentiated enough that it is not simply competing with Amazon and Microsoft on their core turf.
However, the article concludes that a realistic “10x” opportunity from current levels is unlikely given the capital intensity and execution risk already reflected in the story. It suggests a more reasonable outcome, if things go right, would be closer to a 50%–75% return rather than a life-changing multibagger.
Premium gym chains are entering a “golden era” that is ending or already in decline, as rising operating costs collide with shifting consumer preferences toward more flexible, community-based ways to exercise. Long-term memberships are shrinking, margins are pressured by higher rents and facility expenses, and competition from smaller, more personalized…
