Pleias launches small reasoning models optimized for RAG with built-in citations

French AI startup Pleias has released two open-source small reasoning models specifically designed for retrieval-augmented generation (RAG) with native citation support. As reported by Carl Franzen for VentureBeat, the new models—Pleias-RAG-350M and Pleias-RAG-1B—are available under the Apache 2.0 license, allowing commercial use. Despite their small size, the models outperform many larger alternatives on multi-hop reasoning tasks while maintaining strong multilingual capabilities across French, German, Spanish, and Italian. Co-founder Alexander Doria highlighted the models’ development was partly motivated by Europe’s need for efficient, self-hosted AI solutions that comply with regulations like GDPR. The models feature “proto-agentic” capabilities, automatically assessing queries and determining appropriate responses based on available information, with built-in citation functionality that enhances auditability for regulated sectors like healthcare and finance.