MemoryMaker is a straightforward implementation of a Retrieval-Augmented Generation (RAG) application, allowing users to add and retrieve information using vector embeddings. By leveraging OpenAI’s embedding models and FAISS for similarity search, MemoryMaker provides an efficient way to store “memories” and retrieve them based on their similarity to input queries.
Github: MemoryMaker
[Stack: Python, LLM, FAISS, HTML, CSS, JS]
