Verne and Nscale launch 15MW AI Data Centre in Iceland

Verne, the Nordic data centre operator focused on low-carbon infrastructure, has signed a 15MW agreement with AI-specialist hyperscaler Nscale. 

The deployment will be hosted across Verne’s Icelandic campus and is set to become one of the largest liquid-cooled GPU installations in Europe, supporting the growth of sustainable high-performance AI infrastructure across the region.

The project will see approximately 4,600 Nvidia Blackwell Ultra GPUs rolled out throughout 2026. 

The deployment will be 85% liquid-cooled and 15% air-cooled, reflecting a design focus on energy efficiency, compute density and sustainability. 

GPUs will be integrated within Verne’s existing site infrastructure, optimised for AI workloads such as training and inference at scale.

Dominic Ward, CEO of Verne, says: “The pace of change in AI infrastructure is extraordinary. 

“As the demand for GPU capacity accelerates, availability of clean, renewable power has become as important as raw performance. 

“Partnering with Nscale, whose expertise is redefining how AI infrastructure is delivered responsibly at scale, demonstrates how the Nordics are fast becoming a strategic hub for sustainable AI growth.”

Supporting high-density compute with low-carbon power

The selection of Verne’s Icelandic facility reflects growing industry interest in Nordic sites that offer both performance and environmental advantages. 

Iceland is powered by 100% renewable energy, primarily hydro and geothermal and benefits from natural ambient conditions for free cooling. 

These factors allow operators to deliver high-density, energy-intensive AI workloads with a lower environmental footprint.

Verne’s infrastructure is specifically designed for modern AI deployments that require vast amounts of compute power alongside advanced cooling techniques.

By using liquid cooling at scale, the company is able to reduce energy waste and improve overall system performance, while meeting the rising demand from hyperscalers and enterprise users.

Nscale, which operates sovereign-grade AI infrastructure globally, selected Verne based on its track record in deploying sustainable data centre solutions and its access to guaranteed renewable energy.

Scalable infrastructure for AI workloads

The Iceland deployment is part of Verne’s wider Nordic expansion strategy, which includes new campuses in Finland, and early-stage planning activity in France.

These developments are intended to support the continued rise in demand for AI infrastructure, particularly from organisations prioritising energy efficiency, sustainability and data sovereignty.

The use of Nvidia Blackwell Ultra GPUs reflects the growing focus on high-throughput AI infrastructure, with customers looking to scale training and inference workloads efficiently.

These systems require advanced thermal management, high-performance interconnects and sustained power delivery, which Verne’s platform is engineered to provide.

Commenting on the deployment, David Hogan, Vice President Enterprise at Nvidia, says: “The collaboration between Verne and Nscale showcases how NVIDIA technology can enable high-performance AI factories with a focus on energy efficiency and sustainability.

“Deployments like this reflect how organisations are scaling the next generation of AI workloads responsibly, using innovative cooling and renewable-powered data centres."

Reinforcing the Nordics as an AI infrastructure hub

As data centre operators and hyperscalers seek to align performance and sustainability, the Nordic region continues to emerge as a focal point for AI infrastructure investments. 

Verne’s latest partnership demonstrates how energy availability, cooling efficiency and operational agility can be combined to deliver AI-ready capacity with minimal environmental impact.

The 15MW project with Nscale builds on Verne’s reputation for delivering low-carbon, high-performance environments and supports the growing trend of deploying AI infrastructure in locations that offer both operational resilience and regulatory alignment.

With demand for AI compute expected to grow significantly by 2030,  scalable deployments like this one are set to play a central role in shaping the infrastructure behind Europe’s digital and AI ambitions.