“Nano-Banana” Revealed: Google’s New AI Image Tool Arrives

“Nano-Banana” Revealed: Google’s Gemini 2.5 Flash Image Arrives

Written by Aaron S.,

Editor-In-Chief

Last Updated: August 27, 2025

Key Takeaways

Google has released Gemini 2.5 Flash Image, an AI tool that lets users edit visuals using simple text, like adjusting poses or combining images;
The model maintains subject consistency across scenes and can create images using multiple references or diagrams for more accurate results;
Every image includes a hidden SynthID mark and metadata, and the tool is available via Google Cloud, OpenRouter, and fal.ai.

Ace quick missions & earn crypto rewards while gaining real-world Web3 skills. Participate Now! 🔥

BITDEGREE IN YOUR SOCIAL FEED

Google has introduced a new image-focused model called Gemini 2.5 Flash Image, designed to produce and refine visuals through simple text commands.

The update was officially released on August 26 and is already live across all Gemini interfaces.

With this rollout, users can perform detailed edits, such as adjusting a person's stance or combining different visuals, without altering key details like facial features or background elements.

Did you know?

Want to get smarter & wealthier with crypto?

Subscribe - We publish new crypto explainer videos every week!

Before its official reveal, the tool had surfaced on LMArena, a site where the public can test experimental artificial intelligence (AI) models. Known by that name under the alias "nano-banana", the model gained attention for its smooth and reliable editing.

One of the standout features is its ability to keep a consistent appearance for a subject across multiple images. This can be useful in fields like advertising, where a product or character needs to appear the same in different scenes.

The model can also pull from multiple reference visuals and understand diagrams, which allows it to create accurate visuals based on a mix of sources.

To help prevent misuse, all images generated by the model carry an invisible SynthID signature, along with metadata that identifies them as AI-generated.

Gemini 2.5 Flash Image is available not only within Google’s platform but also through OpenRouter and fal.ai. It is also integrated into Google Cloud’s service offering, with a cost of $30 for every million output tokens, which translates to around $0.04 per image.

On August 13, Google rolled out an update to Gemini AI. What’s new in this version? Read the full story.