@kathrynpetrillo.ai's reel — transcript & breakdown

Immediately grabs attention with a bold, controversial claim and a visual representation of complex AI models.

“Someone just ran a 744 billion parameter AI model on a laptop with no graphics card.”

Introduces the protagonist and the specific technology that enables this breakthrough.

“A developer in Italy pulled it off with GLM 5.2, the open model that beats the ones that you pay for.”

Highlights the disruptive nature of this technology by directly challenging a major industry player's premise.

“Nvidia's entire pitch is that you need their GPUs for something like this and you just prove you don't.”

Explains the core technical innovation in an accessible way, comparing it to a common streaming service.

“The project is called Calibri. It works because only about 10 gigs of the model is actually thinking at any moment. So Calibri keeps that part in your RAM and streams the rest off your hard drive as it needs it, the same way Netflix streams movies instead of making you download it.”

Provides a simple, step-by-step guide on how to get the technology working.

“Running it is easier than it sounds. You clone his repo and run one setup file. The whole thing builds itself from there.”

Demonstrates the user-friendly interface and the practical benefits for both direct interaction and app development.

“After that, you type Coli in chat and you're talking to it completely offline. If you build apps, type Coli serve instead and it turns into an Open AI endpoint. You change one line in your code and your token build goes to zero because nothing ever leaves your machine.”

Standard call to action for further engagement, aligning with the video's educational and innovative theme.

“Follow for ingenious engineering and technology explained clearly.”