Google has rolled out a new AI-powered photo editing feature inside Gemini, and Sundar Pichai is marking the launch in his own way. The Google CEO posted three banana emojis on X (formerly Twitter), teasing the arrival of the feature, internally codenamed “Nano Banana.” Officially, it has been launched as Gemini 2.5 Flash Image.
The tool is designed to deliver more natural edits by keeping people, pets and objects consistent across different scenarios. To showcase its ability, Pichai shared a set of AI-generated photos of his dog Jeffree on International Dog Day. The images reimagined Jeffree as a surfer, a cowboy, a superhero and even a chef.
“Our image editing model is now rolling out in @Geminiapp – and yes, it’s 🍌🍌. 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:),” Pichai wrote in his post.
How Google’s Nano Banana (Gemini 2.5 Flash Image) works
According to Google’s blog post, users can upload a photo, describe the changes they want, and Gemini will generate a new version while preserving the original likeness. The tool supports:
Costume and background changes, such as placing a person in different outfits, jobs or eras.
Photo blending, like merging a person and their pet into the same frame.
Step-by-step edits, including adding or removing objects while redesigning a space.
Design mixing, where patterns or textures from one photo can be applied to another.
To ensure transparency, all AI-generated or edited photos carry a visible watermark along with Google’s invisible SynthID digital marker. The Nano Banana feature is available starting today in the Gemini app.