Google Launches "Nano Banana Pro": A Revolution in AI Image Generation and Editing

AI Leadership: Google reinforces its leadership in the world of artificial intelligence by launching its advanced image generation model, "Nano Banana Pro". This new model is a significant upgrade, based on Google's latest large language models, Gemini 3, which was launched on November 18, 2025. This model, also known as "Gemini 3 Pro Image", offers significant improvements over previous versions.

Key Features and Advanced Editing Capabilities of "Nano Banana Pro"

Image Editing Capabilities: "Nano Banana Pro" features advanced image editing capabilities and supports much higher resolutions up to 2K or even 4K, surpassing the previous "Nano Banana" model's maximum of 1024 x 1024 pixels. The new model also provides more accurate text rendering within images, with the ability to generate text in various styles, fonts, and languages with ultra-clarity. Its advanced editing capabilities include consistent character preservation, multi-role editing, and seamless image merging, making it ideal for modifying existing images.

Web Search Capabilities and Professional Image Control

Web Search: One of the most prominent new features in "Nano Banana Pro" is its web search capability, allowing users to accomplish tasks such as searching for specific food recipes and generating flashcards based on them. The model is specifically designed to give professionals unprecedented control over images, enabling them to adjust aspects like camera angles, scene lighting, depth of field, focus, and color grading for precise and customized results.

Cost and Performance: Comparison with Previous Version

High Quality at a Higher Cost: Despite its high quality and superior capabilities, "Nano Banana Pro" is slower and more expensive compared to the original model. While the cost of a single 1024-pixel image in "Nano Banana" was $0.039, the cost for a 1080p or 2K image in the new model reaches $0.139, and it increases to $0.24 for a 4K image.

Advanced Capabilities and User Support

Flexibility in Use: The new model can use six high-resolution shots or combine up to 14 objects within a single image while maintaining consistency and similarity for up to five people. Google has also launched an experimental application that allows users to experience these advanced capabilities themselves.

Availability of "Nano Banana Pro" Across Different Google Platforms

Widespread Adoption: "Nano Banana Pro" is currently being rolled out across many of Google's existing AI tools. The Gemini app will use the new model for image generation by default, with free subscribers able to generate a limited number of images, after which they will revert to the original "Nano Banana" model.

For Subscribers: Subscribers to Google AI Plus, Pro, and Ultra will receive higher generation limits, although the company has not disclosed the exact limits. These subscribers will also have access to the model within Notebook LM. Additionally, the model will be available in Search through the AI mode for AI Pro and Ultra subscribers in the United States. Ultra subscribers can access the model in Google's video tool, Flow, and it is also available to Workspace customers in Google Slides and Vids.

Developer Support and SynthID Technology for Authenticity Assurance

Tools for Developers: For developers, "Nano Banana Pro" can be leveraged via the Gemini API, Google AI Studio, and the new development environment "Antigravity," a new platform designed for developing innovative AI applications. Google is also integrating SynthID technology, its watermarking and AI-generated image detection technology, into the Gemini app. Users will be able to upload any image, and the bot will tell them if the image was created or modified by Google's image models. Google did not mention whether it plans to support other AI-powered watermarking standards such as C2PA, a Content Authenticity and Provenance Coalition that aims to combat misinformation by watermarking digital content.