The European Commission has launched OpenEuroLLM. This initiative develops open-source, multilingual artificial intelligence (AI) models. This project is led by a consortium of 20 European research institutions, businesses, and EuroHPC centers. Jan Hajič (Charles University) and Peter Sarlin (AMD’s Silo AI) lead a consortium which is designed to create Large Language Models (LLMs) supporting all EU languages while abiding with the EU’s regulatory framework.
The project has been awarded the Strategic Technologies for Europe Platform (STEP) Seal, making it the first initiative under the Digital Europe Programme to receive this prestigious recognition. The STEP Seal represents an EU mark of excellence, increasing the project's visibility and improving access to additional EU funding and investment opportunities. As a result, OpenEuroLLM is now featured on the European Commission’s STEP and InvestEU portals.
OpenEuroLLM supports Europe's digital sovereignty through emphasis on ethical, transparent and accessible AI, supporting more balanced representation of languages in AI model training. Unlike many existing models that predominantly train on English-language data, OpenEuroLLM prioritises linguistic diversity, addressing a longstanding gap in the field.
With a total budget of €37.4 million (€20.6 million of which comes from DIGITAL), OpenEuroLLM represents a significant step towards creating AI models which are specific to European linguistic and cultural diversity. By leveraging European supercomputers, the project aims to strengthen the region’s position in AI innovation while ensuring technological sovereignty and maintaining full compliance with the EU AI Act..
Key partners include Aleph Alpha, Germany’s leading AI company; CSC, Finland’s host of one of the world’s most powerful supercomputers; and LightOn, the first European generative AI company listed on the stock exchange. The emphasis of the project’s open-source nature guarantees transparency, with public access to model documentation, training codes, and evaluation metrics upon release. This approach aims to make high-quality AI technologies more accessible so that European businesses can stay competitive in a global market while giving a platform to public institutions to create AI-enabled services.
While specific details on model deployment and applications have yet to be disclosed, continued investment will be fundamental to reduce the technological gap with the current global competitors. As AI continues to reshape industries and societies, OpenEuroLLM stands as an important to Europe’s commitment to technological innovation, transparency, and digital sovereignty. The success of this initiative will not only impact AI development within Europe but also it might also set a precedent for ethical AI practices worldwide.
