SimuLingua

The SimuLingua project aims to deliver an open, multimodal Scientific Foundation Model (SciFM) for materials science that unifies text, chemical/crystal structures, images, and simulation data to accelerate materials discovery.

By fine-tuning existing open-source foundation models with specialized materials data and physics-informed simulations, SimuLingua acts as a “computational linguist” capable of interpreting and connecting diverse “dialects” of materials data.

SimuLingua establishes a closed design–simulate–validate loop:

• The Scientific Foundation Model proposes novel candidate materials.
• Fast AI surrogate models verify their plausibility and performance.
• Validated data flows back into a FAIR knowledge graph.
• A natural-language interface allows non-expert domain scientists to query the knowledge graph, launch virtual experiments, and steer multi-objective inverse design.

Over its 48-month duration, the project will advance the platform from TRL-1 to TRL-4 and validate its capabilities across 6 industrial use cases: lead-free electroceramics, durable glasses, high-entropy and recyclable alloys, analyte-sensitive coatings, and a 64×128 device-level olfactory array.

Consortium

The consortium brings together 9 partners from 7 countries, combining top-tier expertise in AI, computational materials science, physics, experimental validation, and responsible innovation:

• Flowphys AS (Norway) — Project Coordinator
• Eindhoven University of Technology (TUE) (Netherlands)
• KTH Royal Institute of Technology (Sweden)
• Vytautas Magnus University (VDU) (Lithuania)
• NGO Agency of European Innovations (AEI) (Ukraine)
• Dycotec Materials Ltd (United Kingdom)
• Oliveris Tech Incubator (France)
• Thermo Challenges (Sweden)
• The University of Sheffield (United Kingdom)

Role of NGO “Agency of European Innovations” (AEI)

AEI leads Work Packages WP18–20 dedicated to Impact Maximisation (Dissemination, Communication, Exploitation — DEC), driving overall outreach, communication, exploitation strategies, and community engagement.

AEI’s key responsibilities include:

• Communication & Digital Branding: Developing project branding, visual identity, main website, promotional materials, and social media channels.
• Dissemination & Community Engagement: Organizing international and national workshops, webinars, newsletters, conference presentations, and fostering synergy within the GenAI4EU initiative.
• Exploitation & Business Planning: Managing Intellectual Property Rights (IPR) and developing commercialization plans for 5 Key Exploitable Results (KERs), including the SciFM Open-Source Platform.
• Open Science & Governance Support: Contributing to the FAIR Data Management Plan (DMP), Ethical, Legal, and Social Issues (ELSI) compliance, and policy briefs on trustworthy AI.
• Ecosystem Integration: Bridging Ukrainian researchers and innovators with European AI and materials science networks.