in Information Technology

Google’s Nano Banana image generator goes Pro – how it beats the original

18 Views

Getty Images/jastrijebphoto

Follow ZDNET: Add us as a preferred source on Google.

ZDNET’s key takeaways

  • Google released Nano Banana Pro. 
  • The image generator and editor use Gemini 3 Pro. 
  • It is rolling out in the Gemini app to everyone. 

Every now and then, a new model mysteriously appears in the LMArena and creates a stir due to its impressive capabilities — usually a sign that it’s actually worth the hype. Back in August, Google’s Nano Banana image generator made such an appearance, producing hyperrealistic images that left users impressed. Now it’s even more realistic. 

Also: The best AI image generators are getting scary good at things they used to be terrible at

On Thursday, Google launched Nano Banana Pro, built on Gemini 3 Pro, to create visual images that are more accurate representations of your input, utilizing the model’s advanced reasoning and real-time world knowledge. As seen below, it can render more complex infographics with accurate text and information, two tasks that are typically challenging for image generators. 

You can start testing it out today, even if you are not a subscriber. To learn how and why you should, read on. 

Why use Nano Banana Pro?

As the name implies, this model builds upon the model’s previous image-generating capabilities, creating hyper-realistic images that also incorporate real-time information, allowing you to create infographics, diagrams, educational explainers, and more. The information is sourced from Google Search, allowing you to access a wide range of topics related to your images. 

Also: This new most popular AI image and video generator has enterprise users flocking to it

Accurate text 

In addition to gathering the information, the model was designed to incorporate the most accurate and legible text, regardless of length, including long paragraphs. This is a significant advance; less than a year ago, it was difficult to get an image generator to accurately output a single word.

I decided to put it to the test by entering the prompt: 

Can you create an image of a Yorkie that accurately labels each body part in a way that is both accurate and informative, but catered to first graders? Include brief descriptions of what each part is. 

Also: Perplexity’s Comet lands on Android – can it crush Chrome? How to try it

The results were impressive: Not only was the content accurate, but it was also spelled correctly, easy to read, and visually appealing. I’ll let you be the judge…

Generated via Gemini by Sabrina Ortiz/ZDNET

As seen in the image above, the text can also be generated using a variety of textures, fonts, and calligraphy, according to Google. This isn’t limited to English, as the text can either generate content in multiple languages or translate your content into other languages. 

<!–> Google Gemini 3 Pro image generation

–>

Google

More creative options

Sometimes you want to render a new image from existing inputs. Now, you can include up to 12 images with the consistency and resemblance of up to five people. For example, in the image below, the user uploaded 14 different characters, which were then used to create a new image that comprised all of them. 

<!–> Google Gemini 3 Pro image generation

–>

Google

Google has also added more controls that users can access to further personalize an image, including the ability to refine and transform specific parts of the image. For example, Google explains, users can now adjust camera angles, change the focus, apply color grading, and more. 

How to access

The original Nano Banana is still the best option for fast and fun editing, while Nano Banana Pro is ideal for complex compositions, according to Google. To try it yourself, use the Gemini app by clicking on “Create Images” with the “Thinking model.” 

While everyone can access Nano Banana Pro, free-tier users receive limited free quotes, and — upon expiration — will be sent back to the original model. Paid subscribers, including Google AI Plus, Pro, and Ultra users, get higher limits. Google AI Pro and Ultra subscribers will also be able to use it in AI Mode in the US. 

Also: Perplexity’s AI shopping tool is now free for all, just in time for Black Friday – how to try it

While all generations created by Google’s tools bear the SynthID digital watermark, which is invisible to the naked eye, images generated by free and Google AI Pro tier users will continue to include a visible watermark, the Gemini sparkle. A notable update, however, is that now users can upload images into Gemini and ask if it was generated by Google AI. 

Nano Banana Pro is also available to NotebookLM subscribers. Google AI Ultra subscribers will be able to access it in Flow, the company’s AI filmmaking tool. It is also available in the Gemini API, Google AI Studio, Vertex AI, and Google Antigravity to meet the needs of both developers and enterprise users. It will arrive soon at Gemini Enterprise.  

–>