Google has introduced Gemini 2.5 Flash Image, an image generation and editing model that enables features such as blending multiple images into a single image. Developers can use the model for multimodal creativity in visual apps.
The Gemini 2.5 flash images, introduced on August 26th and also identified as “Nano Banana,” allow developers to maintain their characters for consistency, use natural language to perform target transformations, and use Gemini knowledge to generate and edit images. This model is available from Gemini API for developers and Google AI Studio, and Vertex AI for Enterprise. To help you build with Gemini 2.5 Flash Image, Google has updated the build mode for Google AI Studio. According to Google, developers can quickly test the features of their models with custom AI-powered apps, remix them, or realize ideas at a single prompt. The app can be shared from code stored on Google AI Studio or GitHub.
Gemini 2.5 flash images allow targeted transformations and accurate local editing in natural language, Google said. For example, the model blurs the background of the image, removes stains from the t-shirt, removes the entire person from the photo, changes the subject's pose, and adds color to the black and white photo. The key features of the model are:
