Show HN: Raggy – A lightweight CLI tool for RAG over local documents (github.com)

A new lightweight CLI tool named Raggy has been released for Retrieval-Augmented Generation (RAG) over local documents, utilizing technologies like LangChain, Chroma, and Ollama. Raggy can handle multiple common document formats and automatically processes images and scans using OCR. It allows users to generate answers through a local large language model (LLM) or via remote APIs, streamlining workflows for information retrieval from diverse document types. This tool is significant for the AI/ML community as it simplifies the integration of RAG techniques in local environments, enabling developers and researchers to leverage local data without needing extensive setups or cloud dependencies. Raggy’s unique hybrid database combines vector search with a BM25 index for efficient data retrieval, and it can manage changes in document content incrementally, which enhances its usability for dynamic datasets. The detailed configuration options provide flexibility for users to tailor embedding and retrieval processes according to their specific needs, making it a versatile solution for various applications in document management and question-answering systems.