Google Debuts Nano Banana 2.1 Image Generator

Google has introduced Nano Banana 2.1, a new model designed for image generation and editing that replaces the previous Nano Banana 2 model from February 2026. Powered by Gemini 3.6 Flash, the new release operates at Flash-tier speed and supports 1K, 2K, and 4K resolutions. Google positions this release as a highly efficient alternative to its premium Nano Banana Pro model, which currently runs on Gemini 3.1. The company also noted that future quality leaps will likely arrive with the upcoming Gemini 4 Argon Pro model.

For developers and enterprises, the most immediate benefit of Nano Banana 2.1 is a dramatic price reduction. Google has cut generation costs roughly in half compared to the previous version. Generating a 1K image now costs 3.36 cents, down from 6.70 cents, while 4K images have dropped from 15.10 cents to 7.56 cents. This makes the model significantly more affordable than Nano Banana Pro, which charges 13.40 cents per 1K image.

The updated model introduces several functional upgrades for creators. It can process up to 14 reference images simultaneously, maintaining consistency for up to four distinct characters and ten objects across multiple prompts. Additionally, Nano Banana 2.1 connects directly to Google Search and features three selectable thinking levels—minimal, medium, and high—to adjust output quality. Google claims the model improves text rendering, wide panoramas, infographics, and overall visual quality.

While Nano Banana 2.1 occasionally outperforms the Pro version on benchmarks, Google acknowledges that the Pro model still delivers superior realism, color accuracy, and proportions in real-world use. The new model is currently rolling out across Google's ecosystem, including the Gemini app, AI Mode in Google Search, Google AI Studio, Flow, Stitch, Google Ads, and the Gemini Enterprise Platform. Practitioners using older systems should note that Google plans to deprecate the older gemini-3.1-flash-image model on October 29, 2026.