Google has officially lifted the curtain on its much-talked-about AI image editor, codenamed Nano Banana, confirming weeks of speculation among tech enthusiasts. The company announced that the playful-sounding tool is now live under the name Gemini 2.5 Flash Image, integrated directly into the Gemini app for users worldwide.

The buzz around Nano Banana began when unusual but highly consistent AI-edited photos started surfacing online, sparking curiosity about the technology behind them. Late last evening, Google CEO Sundar Pichai put the rumours to rest by sharing a light-hearted post on X (formerly Twitter). His message—three banana emojis—was followed by a set of images showcasing his dog, Jeffree, transformed into various avatars including a surfer, cowboy, superhero, and chef. The edits, shared on International Dog Day, highlighted the model’s ability to maintain Jeffree’s distinct look across different scenarios.

In his post, Pichai wrote: “Our image editing model is now rolling out in @Geminiapp - and yes, it’s (Banana emojis). Top of @lmarena’s image edit leaderboard, it’s especially good at maintaining likeness across different contexts. Check out a few of my dog Jeffree in honour of International Dog Day - though don’t let these fool you, he definitely prefers the couch:).”

How Nano Banana Works

The Gemini 2.5 Flash Image model builds on Google’s earlier experiments with generative visuals. One persistent complaint with AI image generation has been its inability to preserve likeness—whether in pets or people. Nano Banana is designed to solve that challenge by anchoring edits around the original subject’s unique features.

Using the tool is straightforward: users upload a photo, enter a prompt, and let the AI apply edits that stay true to the subject. Beyond single-step modifications, Nano Banana introduces multi-turn editing, which allows people to refine or extend images step by step. Another innovation is design mixing, enabling textures and patterns from one image to be transferred to another. Think of butterfly wings inspiring a dress print or rain boots carrying the texture of flower petals.

The model also supports photo blending, where multiple images can merge seamlessly into one composition. Edited images can even be re-fed into Gemini to generate animated video outputs, expanding creative possibilities.

Global Rollout and Developer Access

The rollout of Nano Banana is not limited to select users—Google confirmed it is available globally through the Gemini app for both free and paid users. To maintain transparency and trust, every AI-generated image will carry Google’s SynthID digital watermark, both visible and invisible, clearly distinguishing them from real photographs.

For businesses and developers, the model is also accessible via the Gemini API, Google AI Studio, and Vertex AI. Pricing has been set at $30 per million output tokens, with each image costing roughly $0.039 to generate.

With its fun branding and robust capabilities, Nano Banana reflects Google’s continued push to make AI both accessible and creative, while addressing a longstanding pain point for image generation: ensuring that edits still feel personal and authentic.