Facilitates Retrieval Augmented Generation (RAG) using LlamaIndex, Gemini embeddings, and LanceDB for local knowledge storage.
Easyrag simplifies the process of building RAG applications by integrating LlamaIndex for document handling, Gemini for generating embeddings, and LanceDB for efficient vector storage. It provides scripts for data ingestion from various sources (code, URLs, PDFs) and a search server for querying the knowledge base, making it easy to create local, embedded RAG systems.
주요 기능
01Leverages LanceDB for vector storage
02Data ingestion from code, URLs, and PDFs
03Uses LlamaIndex for document processing
04Employs Gemini for generating embeddings
05Includes a search server for querying the knowledge base
061 GitHub stars
사용 사례
01Developing a search engine for web content
02Creating a local knowledge base for code repositories
03Building a RAG-based question answering system for PDF documents
