Google has introduced Nano Banana Pro, an upgraded image generation model designed to offer enhanced editing controls, higher resolutions, and improved text rendering. The release, built upon the recently launched Gemini 3 large language model, aims to provide advanced features for image creation across Google's AI ecosystem.
According to the company, Nano Banana Pro represents an advancement over its predecessor, Nano Banana, by supporting more detailed image generation and precise text rendering, including diverse styles, fonts, and languages. The model also incorporates web searching functionalities, enabling tasks such as retrieving recipes and generating flashcards directly from prompts.
Google states the new model is geared towards professional users, providing granular control over aspects such as camera angles, scene lighting, depth of field, focus, and color grading. A key upgrade includes the ability to generate images at 2K or 4K resolutions, a significant increase from Nano Banana's previous cap of 1024 x 1024 pixels. The model also supports blending up to 14 objects within a single image and maintaining consistent appearance for up to five distinct individuals.
While offering higher quality outputs, Nano Banana Pro operates at a slower pace and increased cost compared to the original model. Google specified that generating a 1080p or 2K image with the new model costs $0.139, and a 4K image costs $0.24, contrasting with the Nano Banana's $0.039 per 1024px image. The model is being integrated across various Google AI tools, including the Gemini app, where it will become the default image generation engine, with usage thresholds varying by subscription tier. Google AI Plus, Pro, and Ultra subscribers will receive expanded generation capabilities, while Ultra subscribers will gain access within the company's Flow video tool and Workspace applications like Google Slides and Vids. Developers can access Nano Banana Pro through the Gemini API, Google AI Studio, and the Antigravity IDE.
In addition to generation capabilities, Google has incorporated SynthID, its proprietary technology for watermarking and detecting AI-generated images, into the Gemini app. This feature allows users to verify if an image has been created or modified by Google's AI models. The company also indicated plans to include support for C2PA content credential detection for future content verification.
