Google may soon allow Gemini users to mark up images directly: Report

Google is reportedly testing a new feature that will allow users to mark up images directly inside the Gemini app and on the web. According to a report by Android Authority, the feature, first spotted in earlier code leaks, now appears to be active for some users. The feature, the report says, is visible in both the Gemini mobile app and the desktop version accessed through Chrome. When users attach an image in the Gemini app on Android, a notification appears informing them about the new capability.


The tool lets users draw or highlight specific areas on an image before sending it to Gemini. This makes it easier for the AI to understand which part of the image the user is referring to when asking a question or giving instructions. The feature also supports simple edits based on user input.




Screenshots shared by testers show that Gemini responds accurately when users mark a certain region of an image, helping the AI analyse or modify only that selected portion.


Google has not officially announced the rollout, and it is unclear when the feature will be available to all users.


In a related news, Google launched its latest and most-advanced AI model — Gemini 3 last month. Announcing the new model, Google CEO Sundar Pichai said that the model as “the best model in the world for multimodal understanding.”


Google claims that Gemini 3 is the most-advanced model and it is better at coding. The new AI model is more creative and versatile as compared to the previous iterations. However, the real disruption lies in its distribution strategy:

• Pro version available immediately to users and developers.
• Day-one integration into Google Search, YouTube, and other apps.
• Backed by Google’s full-stack advantage — from research and TPU chips to cloud hosting and consumer-facing platforms.

This end-to-end control means Google can scale Gemini faster and more efficiently than rivals who rely on external partnerships.