ITG NowHome › AI glossary › API pricing
How does AI API pricing work?
API pricing is how AI companies charge developers for using their models, usually by the amount of text sent and received.
- Category Business & money
- Also called price per million tokens, per million tokens
- Updated 20 Oct 2026
API pricing, explained simply
Many AI companies sell access to their models through an API. They charge by usage, not by a flat fee.
The unit is usually the token, a small chunk of text. Prices are often quoted per million tokens.
Input and output are often priced differently, and different models cost different amounts. Check the company's pricing page for current numbers.
On ITG Now, our AI models page and free tool guides explain which tools need paid API access and which are free.
Related terms
TokenAPIAPI keyPay as you goFreemiumRate limit
Questions people ask about API pricing
Do I pay per message or per token?
Usually per token. A longer message or answer uses more tokens, so it costs more.
Why do output tokens often cost more?
Writing an answer takes more work than reading a question. So many companies charge more for output.
Is the API cheaper than a chatbot subscription?
It depends on how much you use. Light use can cost less, heavy use can cost more.