Nano Banana 2 Introduces Next-Level Visuals

Google has unveiled Nano Banana 2, the new AI image generation model officially branded as Gemini 3.1 Flash Image. Optimized for rapid performance and lifelike detail, this model has been integrated as the primary image engine across every Gemini app mode, including Fast, Thinking, and Pro. This latest iteration advances the legacy of the original Nano Banana (launched in August 2025) and its Pro variant, both of which redefined the industry benchmarks for image fidelity.   

With the versatility to generate visuals across various aspect ratios, Nano Banana 2 scales from 512 pixels up to full 4K resolution. This model elevates the standard for professional-tier output by rendering richer textures, dynamic lighting, and significantly sharper details. The model handles complex prompts with greater accuracy and enhances text within images, generating clear, readable typography that can even be automatically translated and localized. These improvements make Nano Banana 2 ideal for marketing, creative projects, and any workflow requiring high-quality AI-generated visuals.

Beyond the Gemini app, Nano Banana 2 is now the default model in Google Search via Google Lens and AI Mode in 141 countries and is integrated into the Flow video editing tool. Google AI Pro and Ultra subscribers continue to have access to Nano Banana Pro for advanced and specialized use cases. Developers can preview the model through Gemini API, Gemini CLI, Vertex AI, AI Studio, and Antigravity. All outputs include a SynthID watermark and are compatible with C2PA Content Credentials, ensuring transparency and authenticity of AI-generated content.