Français
We are committed to ensuring that digital sovereignty is achieved through open source. We made this choice right from the start, and it continues to guide us today in a field where the battle is raging: generative artificial intelligence.
After the LUCIE experiment, we decided to go further, be more open and more committed. Our goal: to develop free, transparent and high-performance technological building blocks so that everyone can build useful, sovereign and ethical AI.
It was with this in mind that OpenRAG was born: a Retrieval-Augmented Generation project. In concrete terms, it is an AI that searches for information from reliable sources to provide more relevant answers while allowing businesses and citizens alike to retain control over their data. The website is already online: open-rag.ai
The press is already talking about it! Thanks to Clément Bohic and the editorial team at Silicon.fr, who are always attentive to developments in generative AI.
In the article ‘Generative AI: LUCIE shelved, LINAGORA tackles RAG’ (22 July 2025), it is emphasised:
"Do we really need a reranker in RAG pipelines?
LINAGORA has enough doubts to explore an alternative mechanism based on the map & reduce approach. The underlying ambition is to develop its own system, called OpenRAG.”
An outside perspective confirms that this battle is not just a technical choice: it is a societal choice. The story is only just beginning.
Read the full article
