Artificial Intelligence Innovation: Google Presents 'Nano Banana' AI Image Editor in Gemini Project
Google Unveils New AI-Powered Photo Editor: Gemini 2.5 Flash Image
Google has announced the launch of its latest AI-powered photo editor, the Gemini 2.5 Flash Image model, in the Gemini app, available in both free and paid versions. The new model, codenamed Nano Banana, is positioning itself as a tool to compete in the AI creativity market, aiming to challenge industry giants like Adobe, OpenAI, and MidJourney.
Google CEO Sundar Pichai showcased the new AI model by editing photos of his dog, Jeffree, in honour of International Dog Day. The edited photos featured Jeffree as a surfer, cowboy, superhero, and chef, demonstrating the model's ability to create diverse and imaginative edits. Despite the transformations, Jeffree remained recognisable in every picture, a testament to the focus of the Gemini 2.5 Flash Image model on maintaining the likeness of the subject.
One of the key features of the new AI model is wide-watermarking on every output, with both visible and invisible tags. This sets it apart from competitors like Adobe Firefly and Midjourney.
The pricing structure for the Gemini app's AI model is cost-effective, with four cents per image or $30 per million output tokens. This could be attractive to startups, designers, and marketing teams in emerging markets experimenting with AI workflows.
Google is making the technology behind the Gemini 2.5 Flash Image model available to developers through its API, AI Studio, and Vertex AI. To access these resources, developers need a Google Cloud account, with the Vertex AI API activated and appropriate roles such as "Vertex AI User" assigned to their service accounts. Initial testing can be done in Google AI Studio, with billing enabled for sustained use, while enterprises can use managed Vertex AI endpoints for scaling.
The Gemini app offers additional features such as stacking edits step by step, remixing textures across photos, blending images into one, and creating short animated clips. The app is globally available for download.
However, no more information is provided about how the AI model works or its official release date. Nor is there any information about any other features or capabilities of the AI model beyond photo editing.
In summary, Google's new AI-powered photo editor, the Gemini 2.5 Flash Image model, offers a cost-effective solution for developers and creators, with a focus on maintaining the likeness of the subject in edited photos. The model's wide-watermarking feature sets it apart from competitors, and its availability through Google's API, AI Studio, and Vertex AI provides a wide range of opportunities for developers to explore and experiment with the technology.
