I gave Google new Nano Banana Pro try, and she immediately took off my clothes. I didn't ask for this, but the AI model apparently decided that my birthday card would look better if it had more skin.
Nano Banana Pro, as the name suggests, is aimed at professionals. Powered by Gemini 3It's essentially an update to the company's popular image creation and editing tool, which has gone viral on social media. turned selfies into hyper-realistic 3D figures. Google says it can create higher-quality images that can be printed, display legible text on images, and combine multiple images into one composition. It's also designed for “people who want to feel like professionals,” said Naina Raisinghani, product manager for Google DeepMind. Edge. Sounds good because I'm by no means a professional. For me, the results were brilliant but confusing. It looked good, but seemed amateurish.
Using the Nano Banana Pro is quite simple: you go to the Gemini app, select “create images” and turn on “thinking” mode. Just plug in the prompt (and an image if you're using one) and get started. It's also free, although there are limitations: quotas are expanded for Google AI Plus, Pro and Ultra subscribers.
Google makes some bold claims, promising “studio-quality design,” “flawless text rendering,” and plenty of elegant and creative edits. To test this, I uploaded a simple photo next to Edges office in New York with the Brooklyn Bridge in the background. I asked the Gemini to switch the lighting from day to night and it worked quite well. The result looks plausible. It even handled details that often confuse image generators, such as whether cars were moving in the right direction. Adjusting the camera angle was just as easy. I asked Gemini to recreate the shot as if it had been taken from a higher angle on the right, and it worked.
Image: The Verge and Image: The Verge/Google, Nano Banana Pro
Google also says the Nano Banana Pro can create infographics and charts to help visualize real-time information, such as weather or sports. Being British, I asked about the weather for the next four days in Washington and New York, where I am currently located. Visually, the infographic would look right at home on a regular forecasting site. The text and numbers looked normal—as opposed to the garbled nonsense often seen in AI-generated images—and at the end, Gemini provided me with a list of quotes that helped me confirm its accuracy.
The model stumbled a bit on more complex tasks. I asked him to summarize the recent Edge the story of how Europe is rolling back its artificial intelligence and privacy laws in comic format. The images and text were indeed executed in a flawless cartoon font, but the comic didn't convey the essence of the story at all, giving only a vague idea of what was happening. Block's Law on Artificial Intelligence instead of. Perhaps the problem was that I gave Gemini a link to the story rather than pasting the text.
Image: The Verge/Google, Nano Banana Pro
When I did this it gave me a passable comic book style summary. It captured the essence of the real story, although I don't think I would have been able to easily understand it if I hadn't written the source material. Phrases were also composed that are not found anywhere in my article.
Image: The Verge/Google
To really feel like a professional designer, I tried my hand at making greeting cards. After all, Christmas is approaching. Considering I only uploaded three selfies, Gemini did a frankly amazing job of creating three full-body versions of me, each in a different outfit and with a different expression. He also created a realistic snowy setting with Christmas trees just like I requested and put on a “Merry Christmas!” from above, as I asked.
Gemini took the liberty when I asked him to change the map's snowy background to an Australian-style summer beach getaway. These freedoms were my fake clothes: two of my clones were topless. It was strange. There were also some prominent Feet created by artificial intelligence and a smiling sandman instead of the snowman from the winter scene (built by my topless doppelgänger). There were problems, however: the sandman had no shadow, unlike other rendered objects in the picture, and the Christmas lights on the palm trees magically glowed in the bright sun. I tested his precision editing skills by asking him to add some muscle to just one clone, which he did in seconds (if only it were that easy in the real world). Overall the quality was excellent and the image would be somewhat believable (minus the abs) if you didn't know that I'm missing a large tattoo on my chest.
Not everything was great though. The model was unable to save exactly the text that I asked for on my card. Instead of “Merry Christmas!” he chose “Australian Summer Christmas!” He also seems to wrestle with animals, with my sister's cat sitting in the same pose on stilts as the reference image I provided in each version of the card (though he was given a fancy Santa hat).
Overall, I was impressed. The Nano Banana Pro is a clear upgrade from the base model. I was able to request more precise edits, and it actually created clear text, removing a huge obstacle that prevents generative AI tools like this from being used in the real world. But, alas, these qualities were not enough to make me a good designer.
