2025-08-27T06:50:02Z

---

Can you imagine a world where your pet can transform into a superhero or a chef, all at the click of a button? Well, Google’s CEO Sundar Pichai just set the stage for that with his latest revelation! Last evening, he took to social media, tweeting a playful trio of bananas, hinting at his company’s involvement in the viral sensation dubbed Nano Banana.

Hours later, Google officially unveiled its groundbreaking AI tool, aptly named the Gemini 2.5 Flash Image. This innovative technology promises to revolutionize how we edit and modify photos, making it easier than ever to bring our wildest imaginations to life.

The Gemini app, which integrates this cutting-edge generative image AI model, is designed to generate high-quality, context-aware images while ensuring the original subjects remain strikingly recognizable. Pichai himself showcased the tool's capabilities with a series of delightful images featuring his dog, Jeffree. In a fitting nod to International Dog Day, he transformed Jeffree into a wave-riding surfer, a cowboy, a superhero, and even a chef. Each image maintained the dog's unique essence while exploring playful themes.

Pichai shared, “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 :)”

So, how does this new Gemini 2.5 Flash Image model actually work? It builds upon Google’s earlier image generation features introduced earlier this year. A common pain point for users has been when AI-edited photos of people or pets don’t accurately resemble them. The Nano Banana update seeks to address that frustration.

Users simply upload a photo and provide a prompt, allowing Gemini’s AI to apply an array of transformations while keeping the subject’s likeness intact. Talk about magic!

Moreover, the tool introduces exciting new features, such as multi-turn editing, enabling users to create scenes step by step. There’s also design mixing, which allows for creative borrowing of patterns and textures from one photo to apply to another. Imagine turning the wings of a butterfly into a chic fabric print for your next dress or giving your rain boots the delicate texture of flower petals!

But wait, there’s more! The Gemini model supports blending multiple photos into one seamless image and allows users to feed their edited images back into the platform to create animated videos. The possibilities are virtually endless!

As part of its global rollout, Google is making the Gemini 2.5 Flash Image model accessible today for both free and paid users. To promote transparency, all AI-generated images will come with a visible watermark along with Google’s invisible SynthID digital watermark, clearly indicating these are AI creations.

For developers and enterprises eager to harness this technology, the new model will also be available via the Gemini API, Google AI Studio, and Vertex AI, priced at $30 per million output tokens, with each image costing around $0.039 to generate.