Course

What AI Costs

Every AI price makes sense once you see the plumbing. This course walks the money end to end: the two billing worlds (subscription vs API), what a token is, why the reply costs more than the prompt, the one formula that estimates any bill, and the levers that cut it — right-sizing the model, caching, batching, reasoning budgets, and the free-vs-paid-vs-local break-even. No pinned prices to go stale — just the method, one live scene per idea.

7 lessonsBeginner41 min

Start the course Get the app

Course lessons

Two billing worlds

Your $20 app and a developer's metered bill are two different economies. Knowing which you're in explains every AI price you'll ever see.

›

What's a token?

Every AI bill is counted in tokens, not words. Once you can see them, prices stop being mysterious.

›

Input vs output

The reply is the expensive half — and every new turn quietly re-pays for the whole conversation. Two facts that explain most AI bills.

›

Estimate a bill

One small formula turns a scary 'per million tokens' price into a number you can predict before you build.

›

Right-size the model

The single biggest way to cut an AI bill isn't a discount — it's not overpaying for a giant model a small one could handle.

›

Why thinking costs more

Reasoning models are smarter on hard problems — but they quietly spend tokens you never see, at the pricier rate.

›

Free, paid, or local?

Subscription, API, or running a model on your own machine — the cheapest choice flips as your usage grows. Here's how to find your break-even.

›

Take the full course

Learn it properly — tracked & personalized

Follow the whole course as a guided journey — sequential lessons, progress, streaks and a personalized path, all in the app.

Get the app