Why is output more expensive than input?

Generating text takes more computation than reading it. With the providers in this comparison, output costs 3 to 5 times as much as input.

Is a subscription cheaper than the API?

For normal chatting, almost always. The API is meant for your own applications and automations, where you pay per use.

What do AI models cost via the API? (price comparison)

Current API prices of GPT, Claude, Gemini and Grok side by side, per million tokens and for realistic monthly usage.

In short

Via the API, Claude Haiku 5.5 is by far the cheapest on 1 November 2026 ($0.10 per million input tokens) and GPT-6 Astra the most expensive ($10). For 200 Dutch-language requests a day you pay from under a dollar to around $75 a month.

Fact-checked on 2 November 2026

This page is built automatically from our model database, with prices as published by the providers. Enter your own usage below.

Requests per dayInput words per request your question + context/documentAnswer words per requestDays per monthLanguage of the text

0 tokens per month

Model	Per request	Per month

Prices in US dollars as published by the providers (excl. VAT), checked on 2026-11-01. Estimate: real token counts differ per model (Claude, for example, counts more tokens per word). Thinking tokens of reasoning models come on top as output.

Free · runs entirely in your browser · nothing is stored or sent