Google has released Gemini 2.0 Flash, a new version of its AI that allows users to edit photos through plain English commands. Unlike other image editing tools, Gemini can understand and modify existing photos while preserving most of the original content. This is accomplished by its multimodal design, processing text and images simultaneously. Users can make requests such as transforming the style of an image or altering elements without needing technical expertise. Tests revealed that Gemini effectively adds or removes elements from photos and can change perspectives, albeit with varying quality. The model is, however, limited in that it avoids making changes involving children or nudity, and it can struggle with artist-specific style applications. While it shows promise for straightforward editing tasks, users may find limitations when it comes to complex modifications, particularly around maintaining, or making specific changes to background elements.

Source 🔗