Copyright © 2025 All Rights Reserved
5 Min
The Lab51 team loves testing out new AI tools. Today, we’re trying out Nano Banana – Google’s newest AI tool, which has made a lot of buzz recently after being proclaimed the “Photoshop Killer.” In this article, we test it out, explore its capabilities, and show you how to use it.
Nano Banana, also known as Gemini 2.5 Flash Image, is Google’s latest AI-powered image editing and generation model integrated into the Gemini app. It was officially introduced on August 26, 2025. Today, there are several options to try out Nano Banana:
One of Nano Banana’s main advantages is that it’s very cost-effective:
Nano Banana is said to be very versatile: One reason it’s hyped right now is the sheer range of capabilities it offers. If you use several AI tools, one Nano Banana could replace many of them. Here are some of its core functions:
Nano Banana solves the problem of complex prompting. Yes, prompt engineering still matters, but communicating with it in the same way we talk to ChatGPT brings this AI tool to a whole new level.
Just type what you want: “Remove the background” or “Make it black and white.” No need for manual layer or brush work!
We asked Nano Banana to change a sheep’s outfit to a cute pink dress and make it hold a flower.
The result? Flawless. That’s basic-level prompt editing, though. ChatGPT can do the same. So let’s see what else Nano Banana can do.
Keeping the same characters across multiple image generations has long been a weak spot in AI tools. Nano Banana fixes this: People, pets, and objects remain consistent through edits.
For instance, it can change someone’s outfit in three different photos or shift the photo angle, and their face stays the same.
We asked it to make a woman sit in an office with a laptop, then pushed further by requesting a side-view shot.
The result made us gasp. Not only did Nano Banana deliver with high precision, but the image didn’t look like AI slop. The output was clean, sharp, and high-quality. Great job, Nano Banana!
Nano Banana seamlessly merges multiple photos into one, and its omnireference accuracy is impressive. Perfect for collages, mockups, or concept art.
We used these images as a reference, and asked for: “A businessman standing next to his green sports car, with a golden retriever by his side and a yellow parrot on his shoulder.”
The result nailed every detail.
Nano Banana can restore photos, colorize black-and-white images, add or remove objects, and sharpen or stylize shots.
We tested it on a stock old black-and-white photo from a family album. We asked it to restore, upscale, denoise, and colorize it.
The result? Jaw-dropping. It flawlessly restored the photo and even added subtle, natural-looking details.
Because Nano Banana is built on Gemini’s world knowledge, it understands references, cultural context, and real-world objects. That means smarter outputs. For example, asking for “a Paris café” actually looks like Paris, not a generic café.
We gave Nano Banana the hardest task. Since it’s connected to Google, we uploaded a Google Maps screenshot and asked it to generate an image of the bridge shown on the map.
And yes, it recreated the exact bridge from the map, truly impressive!
These are just a few of Nano Banana’s capabilities. Its range of functions is incredibly wide.
Nano Banana exceeded our expectations. It’s fast, cost-efficient, and versatile enough to replace several AI tools. We definitely recommend giving it a try.