Google Officially Unveils “Nano Banana” AI Image-Editing Model

Mountain View, CA - August 26, 2025 Google today confirmed that the viral “Nano Banana” image-editing tool is indeed a Google DeepMind project and is being integrated into its Gemini application, marking a significant upgrade to AI-driven visual content creation.

The model, officially named Gemini 2.5 Flash Image, has been lauded for its ability to perform high-fidelity, text-driven edits in near real time while preserving the likeness and style of subjects across multiple transformations. Free Gemini users can now make up to 100 edits per day, and paid subscribers receive ten times that allowance.

In benchmarks on LMArena, Nano Banana surged to the top of image-editing rankings, outperforming competing tools in both speed and consistency of results. Google demonstrated the model’s capabilities by reimagining a reference photo of a young woman as a matador, seamlessly blending garments, backgrounds, and lighting with uncanny accuracy.

Key features of Gemini 2.5 Flash Image include:

• Natural-language editing commands with no need for masks or layers
• Character and object consistency across multi-step edits
• Sub-two-second render times for most transformations

Google has made the tool available immediately to all Gemini users on web and mobile platforms, as well as via the Gemini API, Google AI Studio, and Vertex AI for enterprise developers. The launch underscores Google’s ambition to lead the next wave of AI-powered creative tools.