Google launches Nano Banana 2 image generation model with faster speed

Google DeepMind released Nano Banana 2, a new image generation model that combines capabilities from its previous Nano Banana Pro with faster processing speeds. The model, also called Gemini 3.1 Flash Image, is now available across multiple Google products.

The new model includes advanced world knowledge capabilities that draw from real-time web search information to render specific subjects more accurately. It can generate text within images and translate content for localization purposes.

Nano Banana 2 maintains subject consistency for up to five characters and 14 objects in a single workflow. The model supports various aspect ratios and resolutions from 512 pixels to 4K quality. Google states the model provides enhanced instruction following for complex requests.

The rollout includes the Gemini app, where Nano Banana 2 replaces Nano Banana Pro across Fast, Thinking and Pro models. Google AI Pro and Ultra subscribers retain access to Nano Banana Pro through a menu option. The model is also available in Google Search's AI Mode and Lens feature across 141 countries and territories in eight languages.

Additional availability includes AI Studio, Gemini API, Google Cloud's Vertex AI, and Flow, where it serves as the default image generation model. Google Ads will use Nano Banana for campaign creation suggestions.

Google integrated SynthID technology with C2PA Content Credentials for content verification. The company reported that its SynthID verification feature has been used over 20 million times since November across various languages to identify AI-generated content.

The information is based on a company press release statement.

You May Also Be Interested In

• Alphabet added to Dow Jones Industrial Average, Verizon dropped
• Nokia and AWS expand collaboration on autonomous network platform
• Goldman Sachs Assumes Braze Inc (BRZE) at Buy