Gemini App's Native Image Editing Feature Expands with Context-Preserving Updates

Native Image Editing in Gemini App Now Wider Available

Following the initial announcement last week, native image editing capabilities are now rolling out more broadly in the Gemini app. This feature allows for targeted modifications to generated images and user-uploaded photos through natural language prompts, ensuring that the base image remains intact while specific changes are made.

Key Features:

• Preservation of Context: Natural language prompts can be used to modify backgrounds, styles, replace objects, and add elements like text without altering the core components of the image.
• Visual Watermarking: Google is testing a visible 'ai' watermark in the bottom-right corner of generated images alongside an existing SynthID digital watermark.
• Personal Photo Editing: Users can upload personal photos and request Gemini to generate images with changes, such as different hair colors.
• Integrated Prompts: The app supports multi-step editing that preserves context throughout the conversation, enabling prompts with text and images integrated. Examples include step-by-step instructions or a draft of a bedtime story with accompanying images.

Developer Preview:

Google has also announced a developer preview for Gemini 2.0 Flash Image Generation and editing, available via the Gemini API in Google AI Studio and Vertex AI. Improvements over the experimental version include better visual quality, more accurate text rendering, and significantly reduced filter block rates.