# Tag: retrieval-augmented generation

Retrieval-augmented generation (RAG) is a technique for providing external sources of knowledge to the LLM. It helps improve the accuracy of AI answers by accessing information outside of the model's training data. SillyTavern provides a set of tools for building a multi-purpose knowledge base from a diverse number of sources, as well as using the collected data in LLM prompts.

Chat vectorization searches for messages in your current chat history that seem relevant to your most recent messages. It temporarily shuffles the most relevant messages to the beginning or end of the chat history. This happens when the model's reply to your last message is generated.