Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Poland

Coordination and Support Action · Running

Galaxia OLIGE Galaxia Olige (open Life Science Knowledge Graph Engine) - Europe’s First Explainable Ai Memory Infrastructure for Life Sciences

Coordinator
Smabbler Spolka Z Ograniczona Odpowiedzialnoscia, Poland
Runs
1 July 2026 to 30 June 2027
Status
Running
EU funding
€499k
Total cost
—
Partners
1 organisations in 1 countries
Type
Coordination and Support Action
Grant agreement
101310588

What the project is doing

In the project's own words, from its summary on CORDIS.

Galaxia OLIGE is a breakthrough project to establish Europe’s first Explainable AI Memory Infrastructure for Life Sciences - a sovereign, sustainable, and publicly accessible knowledge system that empowers regulators, researchers, pharmaceutical innovators, and health agencies with transparent, evidence-based intelligence. Galaxia OLIGE introduces a new class of AI infrastructure: a Hypergraph Language Model (HLM) that autonomously transforms unstructured biomedical text into a dynamic, explainable semantic memory layer.

Unlike LLMs, Galaxia does not generate probabilistic output; it performs traceable, deterministic reasoning. Unlike knowledge graphs, it requires no manual schema engineering and supports n-ary biomedical relationships at scale. It operates entirely in-memory on CPU infrastructure - making it energy-efficient, sovereign by design, and aligned with Europe’s Green Deal and digital autonomy goals. This EIC Pre-Accelerator project will advance Galaxia from Technology Readiness Level (TRL) 4-5 to TRL 6 by deploying the technology in a real-world life sciences environment.

Over 12 months, Galaxia OLIGE will integrate the OpenWebSearch.eu Open Web Index as a European data source, process over 200 million characters of scientific and regulatory content, and build the first Public Life Sciences Knowledge Graph for Europe, accessible via both a user-friendly interface and developer API.This platform will enable users to:1. Ask complex natural-language questions (e.g., drug-drug interactions, adverse events, dosage risks, regulatory constraints).2.

Receive deterministic, explainable results with full source provenance3. Trust AI outputs in compliance with the EU AI Act and EMA regulatory frameworks4. Integrate knowledge directly into pharmaceutical R&D, compliance systems, and innovation tools.

Topics

Who is involved

OrganisationCityCountryRole
Smabbler Spolka Z Ograniczona OdpowiedzialnosciaZorawinaPolandCoordinator

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI