Google launches Nano Banana AI, Pichai goes viral with dog photo

Google has officially launched a new AI-powered image editing tool, code-named Nano Banana, now available as the Gemini 2.5 Flash Image. The generative AI model allows users to edit photos with simple prompts while maintaining the likeness of the original subject. To showcase its capabilities, Google CEO Sundar Pichai shared playful edits of his dog Jeffree, highlighting the tool’s versatility and fun potential.

Sundar Pichai Goes Bananas on Social Media

Last evening, Sundar Pichai hinted at the new AI tool by tweeting three banana emojis, confirming Google’s involvement behind the viral Nano Banana sensation. Hours later, Google formally introduced the AI-powered image editor.

To demonstrate the model’s capabilities, Pichai posted a series of images of his dog Jeffree on X (formerly Twitter). In these photos, Jeffree was transformed into various themes: surfing waves, wearing a cowboy hat, turning into a superhero, and even dressing as a chef. Despite the creative edits, the dog’s original likeness remained consistent, showcasing Gemini 2.5’s advanced subject-preservation technology.

Pichai captioned his post:
"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 honor of International Dog Day - though don’t let these fool you, he definitely prefers the couch:)"

How Gemini 2.5 Flash Image Works

The Gemini 2.5 Flash Image model builds upon Google’s earlier AI image generation tools introduced in the Gemini app earlier this year. One of the most common frustrations with AI image editing is losing the subject’s likeness after transformations. Nano Banana addresses this problem by ensuring that edits preserve the appearance of people, pets, or objects in the photo.

Using the tool is simple:

1. Upload a photo into the Gemini app.

2. Provide a prompt describing the desired transformation.

3. AI applies edits while keeping the subject recognisably the same.

Key Features of Gemini 2.5 Flash Image

The new model comes with several advanced features:

• Multi-turn Editing: Users can make iterative changes, building scenes step by step.

• Design Mixing: Borrow patterns, textures, or elements from one image and apply them to another. For instance, butterfly wings can be converted into a fabric print for clothing.

• Photo Blending: Combine multiple photos into a single seamless image.

• Animated Video Creation: Edited images can be fed back into Gemini to generate animated content.

These features make the tool highly versatile for both casual users and professional creators looking to experiment with AI-driven visuals.

Global Availability and Access

Google is rolling out Gemini 2.5 Flash Image globally through the Gemini app for both free and paid users. To maintain transparency, all AI-generated images will carry:

• A visible watermark indicating AI-generated content.

• Google’s invisible SynthID digital watermark for authenticity.

For developers and enterprises, the AI model is also accessible via:

• Gemini API

• Google AI Studio

• Vertex AI

Pricing is set at $30 per million output tokens, with each generated image costing approximately $0.039.

Significance of Nano Banana AI

With Gemini 2.5, Google is addressing the limitations of earlier AI editors that struggled with subject consistency. By combining prompt-based editing, multi-turn changes, design mixing, and animation, the platform offers a powerful yet user-friendly AI solution.

The playful edits shared by Sundar Pichai demonstrate the model’s potential not just for creative experimentation but also for professional uses like marketing, graphic design, and social media content creation.

Conclusion

Google’s Nano Banana AI (Gemini 2.5 Flash Image) represents a major step forward in generative AI photo editing. With easy-to-use tools, multi-turn editing, and global rollout, it allows both casual and professional users to transform images creatively while maintaining subject integrity. By demonstrating the AI’s capabilities through Jeffree the dog, Sundar Pichai highlighted how the tool can make image editing both fun and realistic.