Google is changing model access and Gemini limits for personal accounts starting October 2026

Google has announced changes to model access and limits in the Gemini app for personal accounts. They begin to take effect in October 2026, and for users without a Google AI subscription the changes enter into force on October 9. This is stated on the Gemini Apps support page.

For AI Plus subscribers, Google promises a separate email stating the date from which the changes will affect their account. This concerns personal accounts only.

What's changing with model access

The support page publishes a table of model availability. It lists the Flash-Lite, Flash, and Pro variants and four tiers of Google AI subscription: no plan, AI Plus, AI Pro, and AI Ultra. For each available model you can choose an effort level, from low to high: this improves the quality of answers but uses up the limit faster.

AI Pro and AI Ultra subscribers also have access to Deep Think mode for maximally parallel reasoning.

What changed before

This is not the first adjustment. As of May 17, 2026, the limits and model availability in Gemini Apps changed for users over 18, and as of July 24, 2026, the same rules took effect for users under 18.

Limits in Gemini Apps are calculated by computing load and refresh every five hours until the weekly allowance is exhausted. Usage is affected by the complexity of the request, the features used, and the length of the conversation. Paid users have higher limits: AI Plus gets twice the standard amount, AI Pro four times, and AI Ultra five or twenty times more than AI Pro, depending on the subscription.

Premium models and features consume more resources, among them image, video, and music generation, Deep Research Pro, extended thinking, and Deep Think. With frequent use of such tools, the limit is exhausted faster.

Translation editors: Соловьев В., Пеков А.