Gemini image edits are arriving for more users

Following up on last week’s news, Google is now more broadly rolling out native image editing capabilities within the Gemini app. This is a really important step forward, moving us beyond the previous setup where asking for a modification to a generated image would result in an entirely new picture.

Instead, Gemini now intelligently preserves the core elements of your base image while allowing for specific alterations through natural language prompts. This enhanced control applies not only to images generated by Gemini but also to pictures you upload directly. Want to swap out a background, change the artistic style, replace an object, or even add text? A simple text prompt is all it takes.

As a quick example, I uploaded an image and then asked Gemini to make it black and white while retaining the red color of the notebook. In just a few seconds, I was presented with the following output and I was pretty impressed:

You’ll also start noticing a visual “ai” watermark in the bottom-right corner of generated images. Google is currently testing this addition that complements the already-present invisible SynthID digital watermark, further enhancing transparency around AI-generated content.

Underpinning this advanced functionality is Gemini 2.0 Flash, though it’s important to note that this image editing tool can be invoked regardless of which Gemini model you are currently using. As of today, it looks like this enhanced image editing feature is becoming more widely available across both free and Advanced Gemini accounts in the US. So, go check and see if you have it already, and if you do, get to editing!