Vector database company Qdrant wants RAG to be more cost-effective

Qdrant's BM42 is a new search model that extracts data from chunks of a document which it says makes vector search more efficient.