S4mpl3r/chat-with-pdf

Implements a Retrieval-Augmented-Generation (RAG) pipeline that chunks PDFs, embeds them with Jina AI, stores vectors in ChromaDB, and uses Groq's fast inference LLMs to answer queries against retrieved context. Fully configurable through environment variables and code constants for embedding models, LLM selection, chunk size, and storage paths. Built on LangChain abstractions to orchestrate the embedding, retrieval, and generation workflow.

No commits in the last 6 months.