Google’s Nano Banana Pro integrates text rendering and advanced editing controls

The company has announced Nano Banana Pro, an image generation model that succeeds Nano Banana, launched a few months ago. The new system uses Gemini 3 Pro as its technological foundation and adds real-time information processing functionalities through connection with Google Search.

Among the technical features of the model, the ability to render readable text directly in generated images stands out, both in short fragments and extensive paragraphs. The system allows working with multiple languages and offers content localization and translation options. Integrated text generation includes different typographic styles, textures, and calligraphy.

Nano Banana Pro allows combining up to 14 reference images in a single project and maintaining visual coherence of up to 5 people in a composition. The model offers localized editing tools to modify specific sections of an image, adjust camera angles, modify focus, and apply color grading. It also incorporates controls to transform lighting conditions in generated scenes.

The system generates images with 2K and 4K resolution in various formats. Capabilities include creating infographics, diagrams, mockups, and data visualizations based on information provided by the user or extracted from real-time sources.

Google has implemented SynthID, its digital watermark system, in images generated by Nano Banana Pro. The Gemini application incorporates a verification function that allows uploading images and checking whether they were created with Google AI tools. Images generated by users of free plans and Google AI Pro include a visible watermark, while Google AI Ultra subscribers obtain images without this visible mark, although they maintain the imperceptible digital watermark.

The company plans to extend SynthID verification technology to audio and video content in future updates.

Key points

• Google introduces Nano Banana Pro, its new image generation model based on Gemini 3 Pro
• The system renders readable text directly in images in multiple languages
• Offers localized editing tools to adjust camera angles, focus, and lighting conditions
• Allows combining up to 14 reference images while maintaining visual coherence
• Generates images with 2K and 4K resolution in various formats
• Connects with Google Search to process real-time information
• Incorporates SynthID, digital watermark system to identify AI-generated images
• The Gemini app allows verifying if an image was created with Google AI