Retrieval-Augmented Generation

Retrieval Augmented Generation (RAG) is a model that assists in how responses are generated within the GenAI platform. By combining semantic similarity to a user’s input to an embedded document, the RAG architecture provides a Large Language Model with additional context as chunks of relevant documents and/or data. As opposed to directly injecting user prompts to an LLM, the additional context assists the model in generating and delivering a more reliable and useful response to the user.