Pleias, an AI startup focused on responsible development, has announced a new collection of compact reasoning models engineered for optimal performance in Retrieval-Augmented Generation (RAG) workflows. These models are designed to be readily deployed within search-enhanced assistants, educational tools, and customer support platforms. A key feature is their integrated citation functionality, promoting greater transparency and bolstering user confidence. Pleias emphasizes that these models can significantly enhance the precision and dependability of AI-driven information retrieval systems in diverse sectors.