Nano Banana 2.1 brings 14-reference image generation to general availability

What is changing?

Google made Nano Banana 2.1 generally available on October 6. Built on Gemini 3.6 Flash, the model supports 1K, 2K, and 4K output for text-to-image generation and conversational editing. Google says it improves visual quality, prompt adherence, text rendering, and multi-turn character consistency, while reducing tiling artifacts in wide and panoramic 1:4, 4:1, 1:8, and 8:1 aspect ratios.

The model can fuse as many as 14 reference images in one request, targeting consistency for up to four characters and fidelity for up to 10 objects. It also supports grounding with Google Web and Image Search and configurable minimal, medium, or high thinking levels. Distribution includes the Gemini app, AI Studio, Gemini API, Search AI Mode, Google Ads, Flow, and Stitch.

For production teams, the update could reduce manual rework when recontextualizing characters, products, and campaign assets, while improving text accuracy in posters and infographics. Google’s model card still lists limitations involving small text, long paragraphs, spatial directions, and imperfect character consistency. The previous Nano Banana 2 model is scheduled to shut down on October 29, 2026.