Glossary term
Glossary term
Memory and Retrieval
Basic RAG enhances language models by retrieving relevant information from external sources to generate more accurate, grounded responses, supplementing pre-trained knowledge with real-time or domain-specific context.
A simple LangChain RetrievalQA chain using Pinecone and OpenAI embeddings is a textbook Basic RAG implementation.
Notion AI Q&A retrieves the top relevant pages from a workspace and feeds them to an LLM for grounded answers.
OpenAI's File Search in the Assistants API offers a basic RAG pattern over uploaded documents.