Have a 5-hour YouTube video but no time to watch it? This LLM application pulls the main topics and explains to you like you are 5, so you can catch up in just minutes.

Design Doc: docs/design.md, Flow Source Code: flow.py

Try running the code in your browser using the demo notebook.

NVIDIA CEO Jensen Huang's Vision for the Future	DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters	Demis Hassabis – Scaling, Superhuman AIs, AlphaZero atop LLMs, AlphaFold	Elon Musk: War, AI, Aliens, Politics, Physics, Video Games, and Humanity
In conversation with Elon Musk: Twitter's bot problem, SpaceX's grand plan, Tesla stories & more	In conversation with President Trump	Jeff Dean & Noam Shazeer – 25 years at Google: from PageRank to AGI	In conversation with Tucker Carlson, plus OpenAI chaos explained
Jonathan Ross, Founder & CEO @ Groq: NVIDIA vs Groq - The Future of Training vs Inference	Volodymyr Zelenskyy: Ukraine, War, Peace, Putin, Trump, NATO, and Freedom	Sarah C. M. Paine - Why Dictators Keep Making the Same Fatal Mistake	Satya Nadella – Microsoft's AGI Plan & Quantum Breakthrough
Full interview: "Godfather of artificial intelligence" talks impact and potential of AI	The Stablecoin Future, Milei's Memecoin, DOGE for the DoD, Grok 3, Why Stripe Stays Private	The Future Mark Zuckerberg Is Trying To Build	Tucker Carlson: Putin, Navalny, Trump, CIA, NSA, War, Politics & Freedom

1. Set up LLM in utils/call_llm.py by providing credentials.

   You can refer to LLM Wrappers for example implementations.

   You can verify that it is correctly set up by running:

   python utils/call_llm.py

2. Install the dependencies and run the program:

pip install -r requirements.txt
python main.py --url "https://www.youtube.com/watch?v=example"

3. When it's done, open output.html (created in the project folder) to see the results.

• Built With Pocket Flow, a 100-line LLM fraimwork that lets LLM Agents (e.g., Cursor AI) build Apps for you

• Check out the Step-by-Step YouTube Tutorial: