I tried the Gemini 'nano banana' AI image editing model that topped LMArena
A mystery AI image editor named "nano banana" recently rose to the top of LMArena, the most popular AI leaderboard. The model easily bested its opponents in the arena, which lets users test AI models head-to-head. Now, Google DeepMind has revealed that nano banana is actually the alias for Gemini 2.5 Flash Image.
Before the big reveal, Googlers did drop some hints:
This Tweet is currently unavailable. It might be loading or has been removed.
Now that the model has been officially rolled out, Google DeepMind said that Gemini will be better at editing your images. Products like this move us a step closer to a post-Photoshop world. Instead of learning the technical ins and outs of photo editing software, which can take years to master, AI image editors will make it possible for anyone to edit an image with just a few simple voice or text prompts — in theory.
You May Also Like
Apple AirPods Pro 3 Noise Cancelling Heart Rate Wireless Earbuds — $219.99 (List Price $249.00)
Apple iPad 11" 128GB Wi-Fi Retina Tablet (Blue, 2025 Release) — $274.00 (List Price $349.00)
Amazon Fire HD 10 32GB Tablet (2023 Release, Black) — $69.99 (List Price $139.99)
Sony WH-1000XM5 Wireless Noise Canceling Headphones — $248.00 (List Price $399.99)
Blink Outdoor 4 1080p Security Camera (5-Pack) — $159.99 (List Price $399.99)
Fire TV Stick 4K Streaming Device With Remote (2023 Model) — $24.99 (List Price $49.99)
Shark AV2511AE AI Robot Vacuum With XL Self-Empty Base — $249.99 (List Price $599.00)
Apple Watch Series 11 (GPS, 42mm, S/M Black Sport Band) — $339.00 (List Price $399.00)
WD 6TB My Passport USB 3.0 Portable External Hard Drive — $138.65 (List Price $179.99)
Dell 14 Premium Intel Ultra 7 512GB SSD 16GB RAM 2K Laptop — $999.99 (List Price $1549.99)
The Google DeepMind team says this model has been trained to make subjects more consistent across various edits of AI-generated images. This has been an issue for AI image models, given their unpredictable nature. I tried out the new "nano banana" model for myself, and it worked... fine.
The ability to upload and natively edit photos in Gemini has been around since April of this year. With Gemini's updated model, Google says you can do things like change a subject's outfit and location, while keeping their likeness the same.
You can also upload multiple photos and have the subjects appear together in the same photo, or add and alter specific details in an uploaded image to, say, see what a room looks like with a different color of paint or different furniture.
Here's Gemini's attempt at editing my dog into the downward dog pose and relocating her to a yoga studio. Her likeness is the same, and it successfully edited the image to make her eyes open, but her body isn't arched in the way it should be. (I would know, I've seen this playful pose from her many times.)
As Google DeepMind said in its announcement, the model might not always get it right. There might still be inaccuracies with fine details, text in the image, and inconsistencies. With my experiment, my dog's fur looks overly smooth, but her overall coloring, size, and shape stay the same. All images have a visible watermark and an invisible watermark called SynthID to mitigate any confusion over whether they're real or AI-generated. This update is now live, so you can try it out for yourself in the Gemini app.
Cecily is a tech reporter at Mashable who covers AI, Apple, and emerging tech trends. Before getting her master's degree at Columbia Journalism School, she spent several years working with startups and social impact businesses for Unreasonable Group and B Lab. Before that, she co-founded a startup consulting business for emerging entrepreneurial hubs in South America, Europe, and Asia. You can find her on X at @cecily_mauran.
