Local Llama 3 UI

Chat UI for local offline Llama3 Model to chat with.

Architecture

Streamlit UI

Prerequisites

• Install Ollama - https://ollama.com/
• Install Python 3

Download Llama 3 model

In your terminal:

ollama pull llama3:8b

Create Python Virtual Env

Via conda (miniconda):

  conda create --name local-llm python=3.12
  conda activate local-llm

Or via Python Venv:

  python3.12 -m venv env​
  source env/bin/activate​

​ To deactivate env after the session run:

deactivate​

Install Python libraries

In your terminal:

pip install -r requirements.txt

Run the Streamlit app

streamlit run streamlit_app_v2.py

Create Shell alias (optional)

Add into your bashrc or zshrc file:

alias llama='cd ~/llama3_local; streamlit run streamlit_app_v2.py'

NOTE: update the cd ~/llama3_local with the path, where you've saved this project.

Run the shell alias to call it from any directory

llama