Chat Application using Streamlit and local models

Create conda environment, we'll call it chat

conda create -n chat python=3.11

Install dependencies

pip install openai pip install streamlit

Download Ollama

(Ollama only works on Mac for Linux right now. For Windows you could use LMStudio) https://ollama.com/

Choose and run local model using Ollama

This command will download the mistral model for ollama to use

ollama pull mistral

This command will run mistral at http://localhost:11434/v1

ollama run mistral

Streamlit app

Build Streamlit python file. We'll call it app.py.

This command will run Streamlit app

streamlit run app.py