Nano Banana Pro Is Here: All the Changes in Google’s Popular AI Image Tool
The latest version of Google’s AI image model is here, thanks to Gemini 3. Whether you choose to call it by its official name (Gemini 3 Pro Image) or its better-known nickname (nano banana pro), this model uses Gemini’s “deep thinking” feature to create better content.
Nano Banana Pro promises to be more efficient, especially when it comes to generating readable text, using up to 14 reference images and creating images in resolutions up to 4K.
You can upload up to 14 separate reference images to nano banana pro. Notice how each of the characters in the prompt (left) appears unchanged in the final image (right).
Nano Banana Pro is available now and rolling out globally in the Gemini app. In Gemini, select “Create Images” (with an emoji 🍌) and switch your template to the Thinking template (located on the right side of the prompt window).
You can use nano banana pro for free, but you will reach a build limit faster than paid Google AI subscribers. These plans start at $20 per month. You will automatically revert to the original model once you have used up your credits. The model is also available in Gemini API, Google AI Studio and To flow.
Inside the new nano banana model
AI image generators have notoriously struggled to create clear text; mutilated words are clear signs that an image was created with AI. OpenAI’s GPT-Image-1 was one of the first models to make significant progress in solving this problem, but CNET’s tests showed that it was not capable of reliably creating readable text.
Google says Nano Banana Pro is supposed to be better at this – so good that you can use the new template to create infographics, relying on Gemini’s integration with Google Search (and Search’s AI mode for paid Pro users) to make the process easier.
An example of an infographic created with Nano Banana Pro.
You may have noticed that none of the images made with nano banana pro have Google’s diamond watermark, which generally indicates that they were made with AI. That’s because Google is introducing a new way for its highest-paid subscribers (those who pay $250 per month for the Ultra Package) to remove these watermarks. Even if it is limited to this higher level plane, it is already extremely difficult to say if something was created with AI. Fewer watermarks won’t help. Google says Gemini will be able to identify all images made with its AI thanks to its SynthID, an invisible watermark attached to its AI-generated content. So, in theory, you can upload an image to Gemini and ask if it’s real or made with Google’s AI. We’ll have to wait and see how well this actually works.
Don’t miss any of our unbiased technical content and lab reviews. Add CNET as your preferred Google source.
The original banana nano model, Gemini 2.5 Flash Image, was released this summer. Early testers thought the abilities were…banana. Add to that a series of fruit-themed teasers from Google executives, and the model quickly became better known by its nano banana moniker. Its popularity flew away in the following days.
One of the things fans liked about the original Nano-Banana was how it kept the character consistent. This means that nano banana was able to edit your existing photos without distorting the characters or people in them. CNET’s hands-on testing found this to be true, but there were still some obvious AI slip-ups. The new model could eliminate some of these errors.
It was a big week for Google as it released a major update to its AI system, Gemini 3. The new models are now rolling out, with the aim of giving Gemini users a smarter experience, less sycophantic experience.
To learn more, see how to write the best AI image prompt and the best AI image generators.
