Private AI: Xference launches its European infrastructure from Aruba data centres

Xference has chosen Aruba data centers for their high-power availability per rack, scalability, and operational management of hardware, networking, and security 

A European private AI inference infrastructure designed to ensure users’ digital autonomy: data ownership and sovereignty, top-tier privacy standards, reduced energy consumption, and a secure environment 

London, April 28 2026 – Italian startup Xference announces the launch of its European AI inference infrastructure at the Aruba data center in Ponte San Pietro and officially opens its invite-only beta phase to the first 100 selected early adopters. 

This initiative represents a concrete step toward building a distributed infrastructure across Europe dedicated to private AI inference, designed to ensure data sovereignty, digital autonomy, regulatory compliance, and high energy efficiency. The service enters its beta phase today by invitation, with the first 100 selected users testing the new technology developed in Italy. 

A European infrastructure for private and sovereign AI 

Founded in 2025 in Cagliari, Xference aims to democratise access to artificial intelligence by offering an inference service based on open-source models, executed in private environments where data always remains under customer control. 

To support this vision, the startup selected Aruba S.p.A. as its technology provider for colocation and Hardware-as-a-Service server provisioning, launching its infrastructure operations at the IT3 data center in Ponte San Pietro (Bergamo). With its European infrastructure and long-standing focus on innovation, Aruba strengthens its connection with deep-tech startups, contributing to the development of AI-driven platforms in Europe. 

“Launching our production infrastructure on Aruba marks a key milestone: we are entering the beta phase with our first users, who will be able to leverage the power of generative AI without their data ever leaving their perimeter,” said Andrea Pili, Co-Founder and CEO of Xference. 

“Aruba has demonstrated the ability to fully align with our mission and support us from the earliest stages with responsiveness, per-rack scalability, and an operational model perfectly suited to the needs of a deep-tech startup. This is a first step toward true European digital autonomy.” 

High-density data centers for the next generation of private AI with open-source LLMs 

Xference’s infrastructure requires significantly higher power density levels than industry averages, exceeding 20 kW per rack, along with linear scalability to support future growth. Aruba has provided a solution capable of supporting the startup’s growth trajectory, handling server procurement and deployment based on required configurations, as well as managing networking and security components. 

“The operational launch of Xference’s infrastructure as the first European Inference Provider in our data centers demonstrates how our network of European infrastructures can deliver next-generation, AI-ready systems and support the development of high-performance AI platforms,” said Giancarlo Giacomello, Head of Data Center & Colocation at Aruba.  

“Thanks to the combination of colocation and Hardware-as-a-Service, Xference can scale rapidly by converting capital investments into operational costs, while maintaining the highest standards of security, compliance, and reliability, ensuring full data governance and localisation, security and compliance by design, with energy sourced from renewables.” 

Beta phase launched with the first 100 users 

The invite-only beta phase will involve the first 100 early adopters, who will help test and optimise the service in real-world scenarios. At the end of this phase, Xference plans a gradual and controlled rollout, with the goal of expanding its infrastructure and developing a distributed edge architecture capable of bringing compute nodes closer to end users, further reducing latency and energy consumption. 

“This beta phase marks the beginning of a gradual and sustainable growth path,” Pili concluded.  

“We will monitor workloads in real time to scale the infrastructure efficiently and build, step by step, a leading independent European Inference Provider.”