Skip to content
ONDI

← All terms

AI terms

API (application programming interface)

Japanese API(アプリケーション・プログラミング・インターフェース)

An API (application programming interface) is a set of rules that lets one program use another program's features. AI companies offer APIs so other apps and websites can add AI features and pay for what they use.

An API is a fixed way for one program to ask another program to do work, instead of a screen made for people. It is like an order window at a restaurant: if you order from the menu, food comes out even if you never see the kitchen. AI services such as ChatGPT offer a developer API separate from the app people use, so other companies' apps and websites can add AI features.

Watch out API fees are not a flat monthly price; you pay for what you use. Prices are usually set per 1 million tokens (the units AI uses to count text), with different prices for input and output. The same model may also be retired on a different date in the app and in the API, so if your web company added an AI feature, find out which model it uses.

Source: API - Glossary | MDN · Pricing · Deprecations · Models & Pricing | DeepSeek API Docs (checked 2026-10-26)

See also

A typo here, a clumsy button there.

This site is still being polished. Spot something? Tell us — we’ll fix it and send a small thank-you gift.

🎁 Thank-you gift