Dekode1859/DocLLM

RAG enabled LLM using HuggingFace model - HuggingFaceH4/zephyr-7b-beta

Upload multiple PDFs and chat with an LLM that answers based on the PDF content provided.

FAISS Vectorstore database for content storage and semantic search used as context retrieval system for the LLM to use.

Enables Conversation Memory for the LLM to use and display in the chat window.

Check out the app hosted at HuggingFace Spaces at Dekode/DocLLM

Check out my HuggingFace profile