Google rolls out Nano Banana 2.1 with better image editing and consistency

Google has released Nano Banana 2.1, an upgraded AI model for generating and editing images. It is now rolling out across the Gemini app, Google Search's AI mode, Google AI Studio, Google Ads and the Gemini enterprise platform.

The upgrade focuses on three areas: better visual design with stronger compositions, more precise mask editing so users can tweak parts of an image without redoing the whole picture, and stronger subject consistency, keeping the same person or object looking the same across multiple generations.

For developers, the model ID is gemini-nano-banana-2.1. It supports text, image, audio and video input, and can output images in 1K, 2K and 4K resolution.

OpenAI's ChatGPT Images 2.5 is still the world's leading image generation and editing model, with Microsoft's MAI-Image-2.6 close behind. Competition in this field continues to heat up.