A Retrieval-Augmented Generation (RAG) system that combines semantic and lexical search over research papers using FastAPI, SQLite, and Google's Gemma LLM.
python cors command-line jinja2 cosine bm25 jinja2-templates scratch-implementation gemini-api rag hnsw generative-ai semantic-chunking faiss-vector-database rag-pipeline rag-chatbot gemma3-27b fixed-chunking chunk-overlap
-
Updated
Feb 11, 2026 - Python