I'm learning LLM at the moment, RAG particularly. [1] I ended up built InkChatGPT as my learning project and it was huge fun. It is an AI Agent that could help

I'm learning LLM at the moment, RAG particularly. [1] I ended up built InkChatGPT as my learning project and it was huge fun. It is an AI Agent that could help learning from multiple documents and you can chat with it, thank Chat PDF GPT. I use LangChain as LLM framework to simplify the backend, and using Streamlit as front end UI and deployment. Using OpenAI `gpt-3.5-turbo` model, Use HuggingFace embeddings to generate embeddings for the document chunks with `all-MiniLM-L6-v2` model. To be honest, coming from Mobile development background, learning about ML and reading about LLM models, prompt tuning and various techniques really opens my mind, but the vast information and knowing how to start is difficult. [1] InkChatGPT: https://github.com/vinhnx/InkChatGPT https://github.com/vinhnx/InkChatGPT