This repo will teach you how to:

• Use LLM local or API via Ollama and again via LangChain
• Use Llama 3-8B model
• Build UI with Gradio
• Use case = "Summarize YouTube video using Llama 3"

Run it

Assuming you have the right python environment and other required tools. You can simply run:

python main.py

Tools you will use

• Ollama to run local LLM API
• Llama 3 from Meta, to use as AI brain
• Gradio, to build UI
• pytube a python library for working with YouTube
• LangChain as framework for LLM app
• tiktoken library to estimate token counts

Requirements

• For using this notebook smoothly, we recommend create a python environment based on our provided requirements.txt.
  This can be done by

pip install -r requirements.txt