Google Unveils Nano Banana 2, Bringing Pro-Grade AI Image Creation at Flash Speed to Everyone

Google has unveiled Nano Banana 2, its latest AI image generation model that blends high-fidelity creative capabilities with ultra-fast performance across its ecosystem. The model, technically known as Gemini 3.1 Flash Image, brings advanced features previously limited to high-end image tools to more users and applications.

Nano Banana 2 combines the visual reasoning and quality of Nano Banana Pro with the rapid generation speed of Gemini Flash, enabling users to create and edit images more quickly and accurately.

The model draws on real-time web search information and generative knowledge to render detailed subject matter, infographics, diagrams and localised visual content with improved precision.

It also adds enhanced creative controls such as subject consistency — maintaining the appearance of up to five characters and 14 objects in a workflow — and precise instruction following, better capturing nuances in complex prompts. Users can generate production-ready assets across multiple resolutions and aspect ratios, from 512 px up to 4K, with richer lighting, textures and photorealistic detail.

Nano Banana 2 is rolling out across key products including the Gemini app, Google Search (AI Mode + Lens), Flow, Google Ads, and developer surfaces like AI Studio, Gemini API and Vertex AI.

The rollout aims to make powerful image generation accessible to a wider audience while improving speed, quality and real-world grounding of AI-created visuals.