Nebius Beats Quarterly Revenue Estimates as AI Demand Fuels Growth
Key Takeaways
- •Nebius Group's second-quarter revenue exceeded expectations as demand for AI computing capacity accelerated, with its core AI cloud business growing more than fivefold.
- •The company signed four major contracts averaging more than $1 billion each, while stronger pricing and customer prepayments supported its expansion plans.
- •Nebius raised its 2025 annual recurring revenue outlook to a range of $900 million to $1.1 billion.
- •Nvidia participated in a $700 million private placement in December 2024, and Meta Platforms agreed in early 2025 to a roughly $3 billion contract for Nebius GPU capacity.
- •Upcoming capacity additions include a major expansion of the New Jersey data center and new sites in the United Kingdom and Israel, which will determine how quickly the contract backlog converts into revenue.
Nebius Group reported second-quarter revenue above expectations as demand for AI computing capacity accelerated, with the company's core AI cloud business growing more than fivefold.
The company signed four major contracts worth over $1 billion each on average, while stronger pricing and customer prepayments helped support its expansion plans. In an industry where GPU fleets and hyperscale data centers require heavy upfront capital, long-dated commitments of this scale are what make rapid buildouts financially feasible.
Nebius is an AI infrastructure company headquartered in Amsterdam and listed on Nasdaq under the ticker NBIS. It emerged from Yandex N.V., the Netherlands-based parent company of Russian technology group Yandex, which completed the divestment of its Russian businesses in 2024 and rebranded as Nebius Group to focus on AI-centric cloud computing. Nasdaq trading in the company's shares had been suspended since 2022, following Russia's invasion of Ukraine, and only resumed after the Russian divestment closed.
Led by co-founder and Chief Executive Arkady Volozh, the company builds and operates data centers in Europe and North America that provide GPU-based computing capacity used to train and run large AI models. Chipmaker Nvidia is among the company's investors, having participated in a $700 million private placement in December 2024, and Meta Platforms agreed in early 2025 to a roughly $3 billion contract for Nebius GPU capacity.
Nebius belongs to a cohort of specialist AI cloud providers, often described as "neoclouds," that rent out Nvidia GPU capacity at scale and compete with hyperscalers such as Microsoft, Amazon and Google for AI training and inference workloads. CoreWeave, a peer in the category, went public on Nasdaq in March 2025. Results from these providers are closely watched as a gauge of broader spending on AI infrastructure.
Alongside the quarter, Nebius raised its 2025 annual recurring revenue outlook to between $900 million and $1.1 billion and pointed to upcoming capacity additions, including a major expansion of its New Jersey data center and new sites in the United Kingdom and Israel. How quickly that capacity comes online will help determine whether the company can convert its growing contract backlog into revenue.
Source: CNBC-TV18
