Google is reportedly preparing a major update to its text-to-image generation tool, Nano Banana. The new version, Nano Banana 2, is expected to feature 4K image output and run on the company’s upcoming Gemini 3 Pro Image model, marking a significant step in Google’s generative AI capabilities.

The tool, originally introduced as a lightweight image generator, has steadily evolved within Google’s AI ecosystem. With this new update, Nano Banana 2 appears to be shifting toward a more advanced, professional-grade application for creators and designers.

Also Read| Sustainability in Tech Infrastructure: A Mere Buzzword?

Smarter and Sharper Image Generation

One of the key changes expected in Nano Banana 2 is a multi-step image generation process. Unlike earlier versions that produced images in a single pass, the new model reportedly plans, reviews, and refines visuals before finalising them. This process is designed to detect inconsistencies and improve accuracy, giving users more coherent and realistic results.

Another highlight is improved text and detail rendering. With Gemini 3 Pro Image at its core, the system is likely to produce sharper edges and clearer graphics, particularly useful for promotional designs, posters, and other visually demanding formats.

Google is also said to be developing an “Edit with Gemini” feature, which would allow users to modify existing images directly, adjusting backgrounds, lighting, or clothing without regenerating the image from scratch.

Also Read| Chinese AI Model ‘Kimi K2 Thinking’ Claims to Outperform GPT-5 and Sonnet 4.5 — and It’s Free!

Awaiting Official Launch

Although Google has not yet announced a release date, early previews suggest that the tool is nearing completion. The rollout is expected to be integrated with Google’s Media AI platform, offering creators access to new features and improved performance.