README.md

SFU IT Chatbot w/ Ollama

This chatbot runs locally on your computer! Yay! This RAG Chatbot uses Gemma3-4b LLM and all-MiniLM-L6-v2 for vector embedding.

To run the app, you need Ollama installed which can be found here:
https://ollama.com/download

Then, you need to download Gemma3-4b from your terminal:
ollama pull gemma3:4b

Now, first set up virtual environment:
python3 -m venv venv
source venv/bin/activate

Then install requirements:
pip install -r requirements.txt

Now run the app (Note: This uses about 3-4 GB of RAM):
python app.py