What you need to know
- A tipster discovered evidence in Google’s code that suggests an editing feature is headed for the Gemini app for Nano Banana’s creations.
- Supposedly, this feature will be presented via a pencil icon on images, giving users the chance to “Mark Up” the changes they want by sketching or typing.
- A rumor from a few months ago alleged a similar capability for users through the Gemini app through a “resubmission” option.
There might be new capabilities headed for Google’s popular Nano Banana image generation software in Gemini.
A report from AssembleDebug with Android Authority alleges that Google is working on a way for users to quickly edit Nano Banana’s creations. The tipster suggests that the main steps to get Nano Banana to generate an image remain the same; however, the change is once the result is produced. The post claims the Gemini app will show a pencil icon, akin to what you’d see for editing purposes in other apps.
Upon tapping, users may be taken to a “Mark Up” screen. From there, the tipster shows Google’s progress at developing a “Sketch” and “Text” option, alongside colors for your pencil in the former. It seems that users can circle or highlight the area that requires c a change. After, the image is returned to your chat bar, as it awaits your text description for what Gemini should change.
The tipster’s example reads, “change the color of this to green, as well.” With that, it seems Google’s AI will take your written description and what’s been circled or highlighted into account, so users won’t have to (essentially) repeat themselves. The AI’s image generator (Nano Banana) will take the image, implement the change, and return the result.
The feature seems relatively fleshed out; however, AssembleDebug’s findings didn’t narrow down when users could see this arrive in the app.
A similar discovery
What the tipster recently discovered in a recent Google app version is similar to rumors from last year. Previous reports claim Google was working on an “annotation” feature for the Gemini app. This was also concerning what images the AI creates through its generator. Users could “write or draw” items on their images before downloading them, alongside a possible “resubmission” option.
That resubmission was said to enable users to make changes or additions before adding instructions for the AI to complete. AssembleDebug’s recent discovery is pretty close to this, which could mean Google hasn’t dropped this idea at all and is still actively working on it.
Android Central’s Take
There’s a bit of QoL here with this rumored feature. Right now, if I want to change something, I’ll have to tell the AI and be pretty specific, hoping it’ll understand what I’m talking about. Through this leaked method, users can be specific in the drawing itself; essentially pointing out to the AI that “this area right here needs to be different.” Combining this with your written instructions makes sure the AI gets it right. It’s useful, and will probably help users brainstorm ideas a little more before bringing them to life properly.
