Google has unveiled its AI image-generation model 'Nano Banana 2' (Gemini 3.1 Flash Image), TechCrunch reported on Feb. 26 local time.

'Nano Banana 2' can generate images in various aspect ratios from 512px up to 4K resolution. It can express up to 5 characters and 14 objects. It can also handle complex requests to deliver more vivid lighting, textures and details, the company explained.

With the update, 'Nano Banana 2' will be applied as the default model for all image-generation modes in the Gemini app and for the video editing tool Flow. Through Google Lens and AI Mode, it will be used as the default image generator for Google Search results in 141 countries. Advanced plan users (Google AI Pro and Ultra) can still optionally use 'Nano Banana Pro'.

API support for developers is also expanding. 'Nano Banana 2' will be available in preview via the Gemini API, Gemini CLI, Vertex API, AI Studio and the development tool Antigravity, which launched in November last year. All generated images will include Google's AI-generated image watermark 'SynthID' and will also be compatible with C2PA content credentials, which include participation from Adobe, Microsoft, OpenAI and Meta. Google said the 'SynthID' verification feature has been used more than 20 million times since it was introduced in November.