https://www.geeknetic.es/Guia/3403/Probamos-Nano-Banana-Pro-el-modelo-de-edicion-de-imagen-mas-potente-de-Google.html

Nano Banana Pro is here

Even though everyone expected Nano Banana 2Google has gone off on a tangent with the name of its new image generation and editing model. There is no second version in the strict sense, but rather a suggestive surname: Pro. To make matters worse, “nano banana” is nothing more than an unofficial name, a joke that Google got right. Actually, what I’m talking to you about here is Gemini 3 Pro Image. As you can see, it is a generative image model included within the new Gemini 3 Pro.

And what are your news with respect to the original Nano Banana that we learned about just a few months ago? They are few, but very relevant:

• Readable text in images. One of the improvements is the ability to generate text within images accurately, legibly and in multiple languages.
• Advanced editing and creation. Allows you to generate images in resolutions up to 4K. Offers studio controls to adjust focus, angles, color grading and other settings. You can even combine multiple reference images to create more complex and detailed results.
• Coherence and continuity. Maintains character identity and visual consistency across multiple generations, which is ideal for projects that require continuity.

How to activate Nano Banana Pro on Google Gemini

The first question we must resolve is how to activate Nano Banana Pro on Google Gemini. Today, the company allows everyone to try this new model. It doesn’t matter if you’re not paying for an AI subscription, like Google AI Pro. The only drawback you’ll run into is usage limits. Now, to do some tests, you will have more than enough.

By the way, if Google AI Studio has come to mind, a space where it has always been free to try the company’s latest in AI, I have bad news. It is not possible, for now, to use Nano Banana Pro on this platform, unless you do it through the API, which is paid.

With this in mind, the only way to “play” with Nano Banana Pro is through Gemini, either on your mobile or on the web. For this guide I am going to opt for the second option. However, if you want to try the prompts which I will leave you in this guide from the app for Android or iOS, you can do it without problems.

To activate Nano Banana Pro you must follow these steps:

1. Enter gemini.google.com or in the mobile app.
2. Open the model selector and choose Reasoning. This mode is the one used by Gemini 3 Pro by default.

3. Then click or tap Tools. In the list, select Create image.

   

4. Finally, write the prompt and wait for the image to be generated.

These are the steps you should follow if what you want is to use Nano Banana Pro. If you select the option Fast in the model selector, you will be using the previous version, the original.

Example 1. Generate realistic images

Although Nano Banana came to the fore thanks to its editing capabilities, it is an excellent image-from-text generator. In fact, what is most striking is the realism of their creations, making it extremely difficult to distinguish an image generated by AI from a real one.

For example, I have asked you this:

• Photographic image with some grain. Cold. A selfie in the middle of the Swiss Alps. A couple is seen with a bokeh effect. They are in the middle of a path that takes them to the top of the Matterhorn. The photograph conveys some nostalgia.

The result is this:

As you can see, this is a fairly simple indication, but the photo really conveys that requested nostalgia, including a date that, it must be said, does not coincide too much with the snowy environment that the model has represented. It was a cold autumn, that of 1987.

Other prompt related to the realistic photography that I have sent to Nano Banana Pro has been this:

• A night photo in a narrow street in the Gothic Quarter of Barcelona. The cobblestone floor is wet and reflects the yellow light of the old street lamps. A person passes by walking quickly backwards wearing a dark jacket. The image has contrast, strong shadows and a little grain, as if it were from a mystery movie. In any case, it should be a very realistic photograph.

Things seem to be looking good, right?

Example 2. Text generation

Another novelty of this model is its ability to generate text without errorssomething in which most models failed. This comes in handy, for example, when representing posters within an image. We do not move from Barcelona for this next prompt in Spanish, but with texts in Catalan:

• A front photo of a rustic neighborhood bakery. On the window glass you can clearly read in white painted letters: “El Forn del Barri”. Just below there is a black board on the floor where it says in chalk: “Savoury offer: coffee and croissant.” Blurred bread is seen behind the glass.

Nano Banana Pro performs the task amazingly. This is how it turned out:

Here is a second request for this exercise in which we test text generation with Gemini 3 Pro Image:

• An overhead photo of a messy wooden desk. There’s a yellow post-it stuck to the corner of a closed laptop. On the post-it it is written in blue marker and linked handwriting: “Call mom at 6:00 p.m.” Next to it are some house keys and a steaming cup of tea.

And here the result:

How do you see it?

Example 3. Creation of infographics and documents

The last thing I suggest to you is test text generation within documents, such as infographics. In this case, it is possible to take advantage of the multimodality of the model to send it information, ask it to analyze it and, from it, generate an infographic or a visual document.

Let’s start with this prompt:

• A very clean and minimalist infographic on “How to make matcha tea”. Just three steps with simple drawings and soft green colors. Step 1: A drawing of green powder with the text “Add Matcha”. Step 2: A drawing of pouring hot water with the text “Water at 80°”. Step 3: A drawing of beating with the bamboo brush and the text “Beat well”.

Additionally, it is possible to create detailed documentation from other images, as I have done with this prompt:

• Use the iPhone image I gave you and create an “exploded view” infographic. I want you to disassemble the phone in the air, separating it into several floating layers, one on top of the other, so that you can see everything inside. The topmost layer will be the screen. In the middle, put a layer with the motherboard, the large battery and the camera module. At the bottom, the back cover. Add fine lines and clear text labels pointing out each part: “Display”, “Battery”, “Motherboard”, “Cameras”. The background should be white, like a clean technical document.

Obviously, to achieve this result, I attached an image of an iPhone that I downloaded from a royalty-free repository.

Nano Banana Pro is an amazing model

What Google achieved with its original Nano Banana model was already impressive, but this new model is even better. In fact, I consider that it will be a mine for all those who need to generate graphic resources and want to reduce the time they dedicate to it. Infographics, documents, realistic photos, illustrations… The precision of Nano Banana Pro to understand the instructions and execute the task is its strong point.

As if this were not enough, spawn time has also improved a lot. All images generated in this guide were obtained in just 30 seconds. Of course, the human component is still relevant with this model. I have no doubt that, in the right hands, those of a creative professional, I mean, the potential of this new version of Nano Banana will be enormous.