US-based AI chipmaker Groq has opened its first European data center in Helsinki, Finland, as part of its global expansion efforts. The facility, developed in partnership with digital infrastructure company Equinix, aims to deliver faster, more efficient AI services to customers across Europe.
The new center is designed to support AI inference tasks, offering lower latency, quicker response times, and improved data protection. These benefits are especially important as demand for large language model (LLM) applications continues to grow.
Helsinki was selected for its strong infrastructure and environmental benefits. Finland’s cool climate, reliable electricity grid, and commitment to sustainable energy help reduce operational costs and energy consumption. These factors support the European Union’s push for “sovereign AI” — an initiative that promotes local data processing to enhance privacy and regulatory compliance.
Groq CEO Jonathan Ross said the company made the decision to launch in Helsinki just four weeks before opening, and began handling customer traffic within days. He described this as a demonstration of Groq’s agility and speed.
The facility is powered by Groq’s Language Processing Units (LPUs), chips built specifically for the unique demands of AI language tasks. Unlike traditional GPU systems, LPUs are optimized for faster and more efficient inference, processing models like Llama-2 at over 100 tokens per second while using up to 70% less energy.
This European center builds on Groq’s existing partnership with Equinix, which also includes a data center in Dallas, Texas. With the new location, Equinix customers across Europe, the Middle East, and Africa can now access GroqCloud for high-performance AI workloads.
Equinix’s Nordic Managing Director, Regina Donato Dahlstrom, said the collaboration combines Groq’s AI technology with Equinix’s global infrastructure, offering organizations a strong and secure foundation for AI deployment.
Groq’s expansion comes at a time when global demand for efficient and sustainable AI infrastructure is rapidly increasing, driven by the rise of generative AI and large-scale language models.
