MOUNTAIN VIEW, CALIFORNIA / RankWire.AI / – Google will reduce the Gemini model choices available to free personal accounts from October 9, 2026. Users without a Google AI subscription will have access to Flash-Lite through Gemini Apps. They will no longer receive access to Flash or Pro under the revised structure. Google detailed the changes in updated support information for its consumer artificial intelligence service. The rules cover personal Google accounts rather than accounts managed through workplaces or schools.
Google AI Plus customers will continue to use Flash-Lite and Flash after the changes reach their accounts. However, the subscription will no longer include access to Pro. Google has not announced a single transition date for every AI Plus subscriber. The company says affected customers will receive an email that identifies when their access changes. Subscribers to AI Pro and AI Ultra will continue to have access to Flash-Lite, Flash and Pro.
Flash-Lite is designed as an efficient Gemini model for fast responses and common everyday tasks. Flash provides additional capability while maintaining quicker responses across many types of requests. Pro targets more demanding workloads, including complex mathematics, coding and advanced analysis. Google says the model can process text, files, images and video while applying deeper reasoning. Pro responses can also require more processing time than responses generated by other Gemini models.
Paid Gemini plans keep wider model access
Google is also introducing low, medium and high effort settings for models available under a user’s plan. Higher effort allows a model to spend more resources producing a detailed response. It also consumes more of the usage allowance available to that account. AI Pro and AI Ultra customers will receive access to Deep Think for demanding reasoning tasks. Google describes Deep Think as an option that applies maximum parallel reasoning when handling complex requests.
The new restrictions follow earlier changes to the way Gemini Apps calculates usage limits. Google introduced compute-based limits on May 17, 2026, with allowances refreshing every five hours until weekly limits are reached. The calculation considers factors such as prompt complexity, features used and conversation length. Advanced features and premium models can consume those allowances faster than basic requests. Google later applied the earlier usage-limit changes to users under 18 on July 24.
Google AI subscriptions set different usage limits
People without a Google AI subscription receive the service’s standard usage limits. AI Plus provides twice the standard allowance, while AI Pro increases it to four times the standard level. AI Ultra provides either five times or 20 times the AI Pro allowance, depending on the subscription. Tools including Deep Research, media generation, Pro, extended thinking and Deep Think can use larger portions of those limits. Paid plans therefore provide higher usage allowances alongside expanded model access.
The October changes apply to Gemini Apps used through personal Google accounts and do not govern every Google AI service. Work and school accounts operate under separate access arrangements, while developer services follow their own rules. Users can manage, upgrade, change or cancel Google AI plans through Gemini Apps and Google One. From October 9, free personal users will rely on Flash-Lite instead of Flash or Pro. Google also recommends keeping the Gemini mobile app updated as the revised access structure takes effect.