Skip to content

Google Unveils Gemini 2.5 Flash for Revolutionary Image Editing

Merge images with a single prompt. Edit precisely with text. Google's Gemini 2.5 Flash is here to transform your image editing experience.

In this picture there are paper arts with some sketches.
In this picture there are paper arts with some sketches.

Google Unveils Gemini 2.5 Flash for Revolutionary Image Editing

Google has unveiled Gemini 2.5 Flash, a cutting-edge image editing model now integrated into the Gemini app. This new feature allows users to merge up to three images for photorealistic interior visualizations and complex compositions with a single prompt.

The model, developed by Google, offers several innovative features. It can transfer styles, applying the color, texture, or design of one object to another while retaining its form and details. This enables precise, localized edits via text input, such as blurring backgrounds, removing spots, adding colors, or deleting entire objects without manual selection tools.

Gemini 2.5 Flash also understands and visually represents simple causal relationships in images. It maintains visual consistency of a person, object, or animal across different images, a feature called character consistency. This allows users to merge images while preserving key elements.

The new model outperforms comparable image systems in various image editing tasks. It behaves similarly to GPT-4 from ChatGPT in terms of text understanding compared to lower-text-understanding image models.

Gemini 2.5 Flash is now available within the Gemini app, with users needing to select the 'Flash' language model among the 'AI' image models at the top left. It is also available as a preview version via the Gemini API, Google AI Studio, and Vertex AI, with usage costs of $30 per million output tokens. Google's latest innovation is set to revolutionize image editing and composition.

Read also:

Latest