Google Rolling Out Native AI Image Editing In The Gemini App

After developer testing last month, Google is beginning to roll out native AI image editing to the Gemini app.

This lets you edit images that you manually upload, as well as ones that the Gemini app generated, with natural language text prompts. You can keep asking Google to make edits until you’re satisfied, with context preserved throughout. This includes changing the background, replacing objects, and adding elements.

You should also be able to change styles. In the example below, you see a hat being added to the dog and then changing the grass background to a beach. Another one is how “you can upload a personal photo and prompt Gemini to generate an image of what you’d look like with different hair colors.”

Once the underlying capability is live (there will be an inline prompt), you can use any of the current models to start image editing.

Another benefit to what Google calls “intuitive, multi-step editing” is “richer, more contextual responses to your prompts with text and images integrated.” One example is asking for step-by-step instructions that include images.

For example, you could ask Gemini to create a first draft of a bedtime story about dragons and provide images to go along with the story.

On the safety front, “images created or edited with native image generation will include the invisible SynthID digital watermark.” Google is also “experimenting with adding a visible watermark on all images generated by Gemini.” In the example above, it’s “ai” in a pill-shaped container at the bottom-right corner.

Last month in AI Studio, Google touted how 2.0 Flash excels at rendering images with text regardless of the length, and how it “leverages world knowledge and enhanced reasoning to create the right image.”

Native image editing is gradually rolling out to the Gemini app starting today. It will be “expanding to more people in over 45 languages and most countries in the coming weeks.”