Cost per token
Level 2
What you pay for each chunk of text a model reads and writes.
Term 6 of 6 in Building AI systems
In plain language
AI pricing is usually per million tokens, with different rates for input and output. Long documents and chatty answers both cost more, so prompt length is a budget line.
Think of it like this
A taxi meter that runs on words instead of miles.
Why it matters
Understanding this makes AI cost predictable and shows why summarising before sending is a real optimisation.