A new way to create personalized images in the Gemini app

The Gemini app combines Personal Intelligence, Nano Banana 2, and Google Photos to support personalized image generation. Now, without writing long prompts or directly uploading reference images, you can get more customized results based on context from connected Google apps.

It works with simple prompts alone. For example, if you ask something like "Design my dream home" or "Draw my desert island essentials," Gemini creates an image that reflects the user's interests and tastes.

When you connect Google Photos, even family and pets can be used in image generation. Based on the people and animal labels in your photo library, it can handle more specific requests like "a claymation image of me and my family doing our favorite activities."

Even if the desired result doesn't come out perfectly the first time, control stays with the user.

• If the result is off, you can direct it to regenerate the incorrect parts.
• You can select different reference photos using the + icon.
• You can check which images were automatically selected via the Sources button.
• You can also directly ask Gemini for attribution and source information.

In terms of privacy, Gemini does not directly train its model on your personal Google Photos library. Instead, only limited information, such as Gemini's prompts and responses, is used for training, and connecting Google apps is opt-in, meaning it can be turned off anytime in settings.

This feature will roll out over the coming days to Google AI Plus, Pro, and Ultra subscribers in the US, and will later expand to Gemini in Chrome desktops and more users.