Google has announced a major update to its Gemini app, bringing a next-generation image editing model from Google DeepMind directly into users’ hands. Already recognized as the world’s top-rated image editing AI during early previews, the model is now integrated natively within the Gemini app – giving users more creative control than ever before.
The upgrade focuses heavily on maintaining consistency in likeness when editing photos of people or pets. Whether you’re giving yourself a retro 60’s hairstyle, adding quirky accessories, or creating fun edits with friends, Gemini ensures subtle details stay true to the original subject. This makes it possible to experiment freely without worrying about awkward distortions or mismatched features.
Users can now blend photos together, change backgrounds, or even insert themselves into entirely new settings. From previewing how new wallpaper looks in your room to creating a whimsical portrait of you and your dog on a basketball court, the possibilities are endless. Once edited, images can also be re-uploaded into Gemini to generate short videos, adding another creative layer.
The new editing model supports multi-turn editing, letting you refine results step by step. For example, start with an empty room, paint the walls, then keep going by adding furniture or décor until you’ve built the perfect scene. Gemini also enables style transfer – applying textures, patterns, or colors from one image to another. Imagine designing a dress with butterfly wing patterns or turning flower petals into the design of your rainboots.
Every image edited or generated within the Gemini app includes a visible watermark, along with Google’s invisible SynthID digital watermark, ensuring transparency and responsible use of AI-generated visuals.
The upgraded image editing experience is available starting today for all Gemini app users.