RAG (Retrieval-Augmented Generation)
Retrieval-augmented generation is a pattern in which a language model is given relevant document excerpts at query time — retrieved from a vector store, search index, or knowledge base — and instructed to answer using that material.
Without RAG, a model can only answer from its training data and the prompt. With RAG, the system retrieves passages relevant to the user's question, inserts them into the prompt as context, and asks the model to ground its answer in those passages.
The benefits are factuality (the answer is constrained to the retrieved material), recency (the index can be updated without retraining the model), and citation (the system can show which passage informed which sentence).
Backplain supports RAG through file upload — PDF, JPEG, and Word documents — at every tier, with the same AI Firewall enforcement applied to retrieved content before it is sent to the model.