'Nano Banana': how Google's free AI image generator works

In recent days, the term Nano Banana has become a trend on social networks and tech forums. It is the new model from Google for generating and editing images with artificial intelligence for free.

Integrated into the Gemini app, the system allows users to transform photos with simple text prompts. With this initiative, Google aims to democratize access to digital editing with hyper-realistic results.

What is Nano Banana and how does it work?

The code name refers to the Gemini 2.5 Flash Image model, which specializes in image manipulation. This system preserves the visual coherence and key features of photos, even when significant changes are applied.

The user only needs to upload an image to Gemini and provide instructions in natural language. The model processes the command and returns the modified photo within seconds.

Possible use cases

• "Convert the photo to black and white."
• "Remove the object from the background."
• "Add a clear sky instead of clouds."

These actions, which previously required advanced knowledge in programs like Photoshop, are now within reach of any user.

How to start using it?

Nano Banana is already available at no cost in the app and web version of Gemini. There is no need to download anything extra or configure options: just access and start editing.

The free service has a daily usage limit, although it is sufficient for most users. Meanwhile, for developers or companies, Google's API offers paid access: 30 dollars per million tokens, about 0.029 dollars for each processed image.

Google and the race for generative AI

With this tool, Google competes against OpenAI, Adobe, and Stability AI, which are also investing in models capable of creating and editing realistic images. The difference lies in its direct integration into Gemini, a platform that millions already use daily.