I attempted Google’s ‘nano banana’ AI picture editor that topped LMArena

A thriller AI picture editor named “nano banana” just lately rose to the highest of LMArena, the most well-liked AI leaderboard. The mannequin simply bested its opponents within the enviornment, which lets customers take a look at AI fashions head-to-head. Now, Google DeepMind has revealed that nano banana is definitely the alias for Gemini 2.5 Flash Image.

Earlier than the large reveal, Googlers did drop some hints:

This Tweet is currently unavailable. It might be loading or has been removed.

Now that the mannequin has been formally rolled out, Google DeepMind mentioned that Gemini will probably be higher at modifying your pictures. Merchandise like this transfer us a step nearer to a post-Photoshop world. As an alternative of studying the technical ins and outs of picture modifying software program, which might take years to grasp, AI picture editors will make it potential for anybody to edit a picture with just some easy voice or textual content prompts — in principle.

The Google DeepMind workforce says this mannequin has been skilled to make topics extra constant throughout numerous edits of AI-generated pictures. This has been a problem for AI image models, given their unpredictable nature. I attempted out the brand new “nano banana” mannequin for myself, and it labored… high quality.

The power to add and natively edit photos in Gemini has been round since April of this yr. With Gemini’s updated model, Google says you are able to do issues like change a topic’s outfit and placement, whereas protecting their likeness the identical.

You may also add a number of pictures and have the topics seem collectively in the identical picture, or add and alter particular particulars in an uploaded picture to, say, see what a room appears like with a unique coloration of paint or completely different furnishings.

Here is Gemini’s try at modifying my canine into the downward canine pose and relocating her to a yoga studio. Her likeness is similar, and it efficiently edited the picture to make her eyes open, however her physique is not arched in the way in which it needs to be. (I’d know, I’ve seen this playful pose from her many occasions.)

Here is my canine Lola, not doing yoga
Credit score: Mashable

Here is the Gemini-edited model once I prompted it to open her eyes, put her within the downward canine pose, and alter the background to a yoga studio. It is shut, however not fairly proper.
Credit score: Mashable

As Google DeepMind mentioned in its announcement, the mannequin may not all the time get it proper. There may nonetheless be inaccuracies with high quality particulars, textual content within the picture, and inconsistencies. With my experiment, my canine’s fur appears overly clean, however her general coloring, dimension, and form keep the identical. All pictures have a visual watermark and an invisible watermark referred to as SynthID to mitigate any confusion over whether they’re real or AI-generated. This replace is now stay, so you’ll be able to strive it out for your self within the Gemini app.