intentor / hello-rag-chat

A chatbot that uses a webpage as context for the conversation using RAG on Llama-3

Hello RAG Chat

This application retrieves the contents of a webpage to use it as context when asking questions to a chatbot using Llama-3 LLM.

It was created to demonstrate RAG (Retrieval Augmented Generation), a technique to provide context to an LLM so it can be more accurate when generating results.

Inspired by the article How to Build a Local RAG App with Llama 3 (Complete Guide).

Setup

Download and install Ollama, a framework to interact with LLMs.

After installing Ollama, you have to download the Llama-3 LLM:

ollama pull llama3

In order to use a language model, you have to start Ollama:

ollama serve
# or
make serve

In a new terminal, run the command below to setup the Python virtual envrinment and dependencies:

make setup

Running the app

make
# or
make run

The chat page will open in your browser. You can enter a URL to generate context for the conversation and start asking questions about it after loading.