Google Gemini’s new image editor keeps characters consistent, fuses images and understands sketches | Technology News

Google Gemini may not be the most popular AI chatbot, but it still has millions of users worldwide. Earlier this year, in April, the tech giant added native image editing capabilities to the Gemini app. Now, Google is giving Gemini’s image editing feature a significant boost.

In a blog post, Google announced that Gemini’s image editing feature is now powered by Gemini 2.5 Flash Image large language model, which can now “blend multiple images into a single image, maintain character consistency for rich storytelling, make targeted transformations using natural language, and use Gemini’s world knowledge to generate and edit images.”

What’s new with Gemini’s image editor?

This means you can now use images generated by Gemini and edit them while focusing on maintaining a character’s traits and details from “one image to the next.” For example, the appearance of a character or object can now be maintained across multiple prompts and edits, place it in different environments, and show a product from multiple angles while preserving the subject.

Story continues below this ad

Google says that its new model can also adhere to visual templates like real estate listing cards, dynamic product mockups for a catalogue and uniform employee badges. Gemini 2.5 Flash Image also supports “precise local edits with natural language” to do things like blur the background of an image, erase a person, remove a stain on a shirt, and add colour to a black and white photo.

The tech giant added that the new text-to-image generation large language model also has “native world knowledge.” This means Gemini can now read and understand hand-drawn diagrams, solve real-world questions and follow complicated editing instructions all in a single step. Gemini 2.5 Flash Image can now also lets you put an object into a scene, fuse images, apply a texture or put a colour scheme in a room.

Source link