OpenAI and Google limit free access to Sora, Gemini and Nano Banana Pro
OpenAI and Google have discreetly implemented new limitations on the free use of their artificial intelligence tools, Sora, Nano Banana Pro and Gemini 3 Pro, starting November 28, 2025. The decision directly impacts the routine of users, who will now notice fewer notifications and a reduction in the frequency of videos and images generated.
The companies have reduced the number of Sora video generations to six per day for free users. Google, in turn, changed the rules for the Gemini 3 Pro in the free version and lowered the limit for the Nano Banana Pro.
These changes mean more restricted access than previously available, affecting users’ ability to interact with these platforms.
On the same topic: Teenager kills mother and brother after using AI to create fantasies about the crime
OpenAI now limits Sora 2, its AI video creation tool, to a maximum of six daily generations for free accounts. Google removed the five suggestions per day guarantee on the Gemini 3 Pro and fixed imaging with the Nano Banana Pro at just two per day for non-paying users.
Rising GPU Demand and Costs Raise Barriers
The main justification for the new restrictions is the significant increase in demand for these services and the rise in costs of Graphics Processing Units (GPUs), which are essential for the operation of AIs.
“Our GPUs are overheating and we want to give as many people as possible access to Sora!” said Bill Peebles, head of the Sora project at OpenAI, when announcing the measure. He added: “We are limiting free usage to 6 generations per day.”
The expression “GPUs melting” highlights the gigantic computational demands of videos created by Sora. Video and image generation consumes significantly more energy than text, as it involves the complex processing of millions of pixels and multiple layers of information for each frame, in contrast to mere sequences of characters. Therefore, limiting the number of free videos aims to alleviate the load on OpenAI’s servers.
Users of paid plans such as ChatGPT Plus and Pro also face their own limits, although Peebles said it is possible to purchase additional generations. However, the purchase option is not attractive for those who rely on free services.
See Also
Electrical discharge knocks man off roof in Jaguaruana after contact with power lines
Global auction raises US$11.9 million and Brazil’s historic currency reaches record value
Former fighter Prichard Colón dies at age 33 after decade in a vegetative state
More on this story: Google introduces Pixel 11 series with improved Tensor G6 chip and new Pixel Tag
What has changed in free access to Gemini 3 Pro
Recently launched for all Gemini users, Gemini 3 Pro initially allowed up to five daily suggestions for those using the free version, in addition to three images per day using the Nano Banana Pro generator. This offer was similar to the previous model, Gemini 2.5 Pro.
However, Google reported, through a support article, that the prompt allocation for Gemini Pro has been modified. The system no longer specifies a fixed number of daily suggestions, only indicating that free users will have “basic access”, with limits that may vary. Creating and editing images with the Nano Banana Pro is now restricted to two per day.
For subscribers to Google AI plans, such as Google AI Pro and Google AI Ultra, nothing has changed, maintaining the limits of up to 100 and 500 daily suggestions, respectively.
A possible indication of changes in the AI business model
The restrictions implemented simultaneously by Google and OpenAI may not be just a coincidence, but indicative of a maturing artificial intelligence industry. This scenario reveals the pressure of extremely high computational costs, especially for generating complex images and videos.
Learn more: Battlefield 6 releases free access for PS5, Xbox and PC; see how to play and when to enjoy multiplayer
It is likely that in the near future, more “incentives” will emerge for users to opt for paid plans. AI companies’ business models are adjusting to the reality of the high operating costs of these technologies.
With demand growing rapidly, GPUs in constant demand, and AI companies subtly steering users towards paid plans, the market is changing. While AI tools are not likely to disappear, free and generous access to the most advanced models may be numbered.
