Nebius to invest more than USD 1 billion to build AI infrastructure in Europe

Build-out includes expansion of existing facilities; additional capacity deployment through colocations; and new-build GPU clusters at greenfield sites.

  • 8 months ago Posted in

Nebius has launched a new GPU cluster in Paris, as part of the company’s plans to invest more than USD 1 billion by mid-2025 in AI infrastructure in Europe.

Arkady Volozh, founder and CEO of Nebius, said: “We work in a new industry which requires both deep technology and significant capital. Our data center in Finland already provides the latest high-performance compute, tools and services to AI developers around the world. The addition of our new GPU cluster in Paris is the next step in our plan to expand Europe’s AI capacity as we develop Nebius into a leading global AI infrastructure company.”

Nebius’s new GPU cluster in Paris will be among the first in Europe to offer NVIDIA H200 Tensor Core GPUs. A cloud and original equipment manufacturer partner in the NVIDIA Partner Network, Nebius will also be one of the first to bring the state-of-the-art, energy efficient NVIDIA Blackwell platform to customers in 2025.

Already one of the leading providers of GPU capacity in Europe through its highly energy-efficient data center in Mäntsälä, Finland, Nebius follows an AI infrastructure build-out strategy that combines investments in new-build data centers at greenfield sites primarily in Europe with colocations and the addition of capacity at its existing facilities. In Europe, Nebius has signed letters of intent to build two new data centers. The company has also started work on expanding its data center in Finland.

Together, these investments – which include approximately 200 million USD deployed since the beginning of this year – will enable the company to reach a total capacity of tens of thousands of NVIDIA GPUs to bring its highly differentiated, energy efficient, complete AI cloud offering to customers worldwide.

Nebius is being purpose-built to meet the demands of the global AI industry and leans on deep technical expertise across hardware, software and machine learning. The company’s 500+ strong team of engineers has decades of knowledge of building world-class tech infrastructure as well as an in-house LLM R&D team.

58% of organizations have observability programs, but 42% still don’t trust their AI models.
GenAI is expected to improve productivity, efficiency, and automation, with long-term ROI for ...
Survey of 500 tech leaders shows companies are moving fast on AI plans while struggling with scale,...
New industry report highlights growing SaaS waste, persisting audit costs and evolving priorities...
LTIMindtree has launched a new business unit and suite of AI services and solutions: BlueVerse.
New Endava research shows 93% want rapid adoption – even as infrastructure struggles to keep up.
According to a new report from Ericsson, 87 percent of business leaders surveyed in Europe believe...
New research from Cloudsmith found 42% of developers using AI in their workflows say at least half...