Google is testing a new feature for its Gemini app and web. Users can now mark specific areas on images before sending them to Gemini. This helps the AI understand requests better. The tool also supports simple edits. This comes as Google recently launched its advanced Gemini 3 AI model.
