AI APIs

How much does the Gemini API cost?

Gemini API prices per model checked in November 2026, what the free tier gives and costs you in data use, and how to keep a Gemini feature within budget.

The Gemini API has a free tier for many models, with rate limits and the condition that free-tier content may be used to improve Google's products. Paid use is billed per million tokens. On 9 November 2026, Gemini 3.1 Flash-Lite cost 0.25 dollars input and 1.50 output, and Gemini 3.5 Flash cost 1.50 and 9 dollars.

Published 9 November 2026 · Last updated 9 November 2026

By Michal Měřínský. Audit methodology.

Free tier

Many Gemini models can be called free within rate limits, using a key from Google AI Studio. Google's pricing page states that free-tier content may be used to improve its products, and that paid-tier content is not. Some models, such as Gemini 3.1 Pro Preview, are not available on the free tier.

Do not send customer or personal data through the free tier. Enable billing on the project for production.

Paid prices per model

Google's pricing page on 9 November 2026, per million text tokens, input then output:

Gemini 3.1 Pro Preview: 2 dollars and 12 dollars for prompts up to 200,000 tokens, 4 and 18 dollars above that.
Gemini 3.5 Flash: 1.50 dollars and 9 dollars.
Gemini 3 Flash Preview: 0.50 dollars and 3 dollars.
Gemini 3.1 Flash-Lite: 0.25 dollars and 1.50 dollars.
Gemini 2.5 Pro: 1.25 dollars and 10 dollars up to 200,000 tokens.
Gemini 2.5 Flash: 0.30 dollars and 2.50 dollars.
Gemini 2.5 Flash-Lite: 0.10 dollars and 0.40 dollars. Audio, image and video input can be priced differently. Preview models can change or be withdrawn, so check the page before committing.

Keeping costs predictable

Prototype on the free tier, then measure tokens per request before choosing a paid model.
Use Flash-Lite or Flash for high-volume tasks, and Pro only where quality demands it.
Set budgets and alerts on the Google Cloud billing account.
Limit output tokens and cache results that do not need regenerating.
Rate limit your own endpoints per user.

Estimating a feature's cost

A request with 2,000 input tokens and 500 output tokens on Gemini 2.5 Flash costs about 0.0006 plus 0.00125 dollars. Ten thousand a day is under 20 dollars. The same volume on a Pro model costs several times more, so model choice is the biggest lever.

What a scan can and cannot tell you

WebAuditOS checks the parts of an AI-powered site visible from outside: CORS on your endpoints, HTML error pages with recognizable debug output, and third-party script load.

It cannot see your Gemini usage, billing or keys.

A free-tier prototype that reaches production with real users is both a privacy problem and a reliability problem. Move to billing before launch.

Short answers

Is the Gemini API free? Many models have a free tier with limits. Production use is usually paid.

Does Google train on my API data? Google's pricing page says free-tier content may be used to improve products. Paid-tier content is not.

What is the cheapest Gemini model? Among current models listed on 9 November 2026, Gemini 2.5 Flash-Lite.

Where do I set a budget? In the Google Cloud billing account linked to your project.

In the product

The checks this maps to

Every check named here runs on every scan, on every plan - what a plan changes is how many pages are crawled, how long the history is kept and what you can do with the result. These are the ids the findings carry, so you can match a line in a report to the explanation above.

sec.cors.wildcard - CORS is not wide open
sec.error-disclosure - Error pages do not leak stack traces
sec.third-party-scripts - Third-party script count is controlled

Run a free audit See a full sample report See the whole check library