As part of the Apply AI Strategy, the Commission, together with the European High Performance Computing Joint Undertaking (EuroHPC JU), has launched the Frontier AI Grand Challenge, a flagship EU-wide competition to drive the development of sovereign, large-scale European AI models. The initiative aims to close the strategic gap in high-end AI by supporting the creation of “frontier” general-purpose AI systems capable of adapting across domains with minimal modification, and by leveraging Europe’s world-class supercomputing infrastructure.
The challenge invites leading European actors to develop frontier AI models with a computational capacity equivalent to at least 400 billion parameters. By using efficient, modular architectures like Mixture-of-Experts (MoE), these models are expected to set new benchmarks for performance and efficiency. This initiative complements wider EU efforts to make Europe the AI continent, as set out in the AI Continent Action Plan, and to strengthen Europe’s AI startup and scale-up ecosystem, ensuring that the most promising innovators have access to the infrastructure needed to compete globally.
Implemented under the EU-funded AI-BOOST project, the Frontier AI Grand Challenge will select one proposal to train a frontier AI model designed to outperform state-of-the-art systems on a range of relevant tasks, using EuroHPC computing resources. The selected project will receive up to 2.5% of the overall EuroHPC computing capacity for one year on one or more AI-optimised EuroHPC supercomputers.
The Frontier AI Grand Challenge is expected to deliver open models that will be made widely available to public authorities, scientific communities and businesses across Europe, supporting innovation in key ApplyAI sectors such as manufacturing, healthcare and autonomous systems. By combining world-class EuroHPC capacity with Europe’s leading AI talent, the initiative will reinforce the EU’s position in global AI competition and foster a vibrant, trustworthy and human-centric AI ecosystem “Made in Europe”.