Erasmus+ KA220-HED · 2026 – 2029

Helping educators build AI that answers from trusted course materials

RAG4Edu is a European project for higher education. We help educators and students design their own AI assistants with Retrieval-Augmented Generation (RAG) — through open training, simple tools and a community of practice.

1What is RAG4Edu?

A European project on AI for teaching and learning

Universities and AI research centres from across Europe work together to help educators use generative AI in a way that is grounded, transparent and responsible.

Our key objectives

  1. Build educators’ skillsHelp educators understand RAG and use it in their teaching — no advanced AI skills needed.
  2. Create open toolsDevelop simple tools and templates for course-specific AI assistants on European language models.
  3. Test it in real coursesPilot and improve a training programme with universities in the partner countries.
  4. Grow a European communityConnect educators across Europe and share all results under open licences.
RAG4Edu illustration: a teacher and four students at a classroom table with a friendly AI assistant and a RAG4Edu laptop. A student asks “How does climate change affect biodiversity?”. The assistant retrieves textbooks, academic articles, lesson materials and trusted databases, then generates an answer that is grounded in sources, provides citations and is trustworthy. Bottom line: students, teachers, AI plus trusted knowledge, deeper understanding, a more inclusive, informed world.

2Why does it matter?

AI in teaching should be reliable — and show its sources

General chatbots answer from whatever they learned online: their answers can be wrong, and you cannot see where they come from. With RAG, an AI assistant answers from the materials an educator chooses, and shows them.

  1. 1Retrieve

    A question is matched with the course materials the educator selected.

  2. 2Augment

    The most relevant passages are added to the question.

  3. 3Generate

    A European language model writes the answer and cites its sources.

  4. 4For education

    Educators design and check every step, with data protection built in.

What educators told us

Most educators in our needs analysis would like to use RAG in their teaching, but do not yet know how it works. RAG4Edu is designed to close that gap.

Read the needs analysis findings →

3What can I learn or do?

Learn, build and share

Training, tools and resources are developed and piloted step by step during the project, and everything is published under open licences.

Learn

Take the training

Short online modules, group activities and micro-credentials help you learn how to use RAG in your own courses, at your own pace.

About the training →
Build

Create a course assistant

The RAG4Edu Toolbox, templates and worked examples help you set up an AI assistant for your course without advanced AI skills.

How it works →
Reuse

Use open resources

Reports, the RAG4Edu Curriculum, tools and templates are free to reuse and adapt in your own institution.

Results & resources →

4How can I get involved?

Join the RAG4Edu Community of Practice

Meet educators, teaching and learning centres and researchers from across Europe in webinars, co-design sessions and roundtables, and help shape the training in the pilots.

One free RAG4Edu account gives you the members’ area of this website and the courses on the RAG4Edu learning platform.

A partner presents the work plan on a projector screen while colleagues work on laptops around a meeting table. A partner waves while the group walks with their luggage along a tree-lined street in Berlin.

Latest news

All news

Upcoming events