Google Nano Banana Pro: New Image Generation Model Released

Google Enhances Image Generation with Nano Banana Pro

Google is implementing upgrades to its image-generation capabilities, introducing advanced editing features, increased resolution options, improved text rendering accuracy, and integrated web search functionality.

Introducing Nano Banana Pro

The newly developed model, named Nano Banana Pro, leverages Google’s most recent large language model, Gemini 3, which was launched earlier this week. Google asserts that Nano Banana Pro represents a significant advancement over its predecessor, Nano Banana, offering enhanced detail in image creation, more precise text representation, and the ability to produce text in a variety of styles, fonts, and languages.

This model incorporates web-searching capabilities, enabling tasks such as requesting a recipe and subsequently generating corresponding flashcards.

Expanded Control and Resolution

Google emphasizes that Nano Banana Pro is designed to provide professionals with greater control over image characteristics. Users can now manipulate elements like camera perspectives, illumination within the scene, depth of field, focal points, and color grading.

Furthermore, Nano Banana Pro surpasses the resolution limitations of Nano Banana, which was capped at 1024 x 1024px. Users can now generate images in 2K or 4K resolutions.

Cost and Performance Considerations

While Nano Banana Pro delivers superior image quality, it comes with increased computational demands and associated costs. The original Nano Banana model was priced at $0.039 per 1024px image. In contrast, the new model costs $0.139 for each 1080p or 2K image, and $0.24 for every 4K image.

Advanced Image Manipulation

The new model supports the utilization of up to six high-fidelity source images or the integration of as many as 14 distinct objects within a single image. It also maintains consistent visual characteristics and resemblance for up to five individuals.

A demonstration application has been released, allowing users to explore these new functionalities.

Rollout and Subscription Tiers

Nano Banana Pro is being integrated into numerous existing Google AI tools. The Gemini application will now utilize the new model for image generation by default.

Users with a free subscription will have access to the model for a limited number of image creations, after which they will revert to the original Nano Banana model. Google AI Plus, Pro, and Ultra subscribers will receive higher generation limits, though specific thresholds have not been disclosed.

These premium subscribers will also gain access to the model within NotebookLM.

Availability in Search and Workspace

Google is also integrating the model into its search functionality through AI mode, available to AI Pro and Ultra subscribers in the U.S. Ultra subscribers can utilize the model within Flow, Google’s video tool.

The model is also accessible to Workspace customers within Google Slides and Vids.

Developer Access

Developers can access Nano Banana Pro through the Gemini API, Google AI Studio, and the company’s newly introduced IDE, Antigravity.

AI-Generated Content Detection

Google is incorporating SynthID, its technology for watermarking and identifying AI-generated images, into the Gemini app. Users can upload an image, and the chatbot will determine if it was created or modified by Google’s image models.

The company plans to add support for C2PA content credential detection for enhanced content verification in the future.

This article has been updated to reflect Google’s plans regarding C2PA support.