Google has announced Nano Banana 2, its latest image generation and editing model. Officially designated as Gemini 3.1 Flash Image, the model is designed to provide professional-level visual quality and advanced reasoning at the high speeds associated with Google’s Flash architecture. It succeeds the original Nano Banana (Gemini 2.5 Flash Image) and integrates features previously exclusive to Nano Banana Pro.
In their announcement, they share that a key feature of Nano Banana 2 is its advanced world knowledge, which leverages Gemini’s knowledge base and real-time information from Google Search to render specific subjects, current events, and niche technical details with higher accuracy. The model will also have significant improvements in precision text rendering and translation.
Nano Banana 2 is production-ready, supporting a wide range of aspect ratios and resolutions from 512px up to 4K. It is currently rolling out across the Google ecosystem, including the Gemini app, Google Search, Google Ads, and the Flow video tool. Developers can also access the model in preview via the Gemini API in Vertex AI and Google AI Studio.
