Back to all articles
•Technology

Google Enhances Image Generation with Launch of Nano Banana 2

View original source

Google announced the release of Nano Banana 2, the latest update to its image generation model, technically named Gemini 3.1 Flash Image.

  • Functionality and Features: This new model surpasses its predecessor by creating more realistic images swiftly and retaining key characteristics of the Nano Banana Pro. Users can generate images ranging from 512px to 4K, maintain character consistency with up to five characters, and fidelity for up to 14 objects.
  • Integration: Nano Banana 2 will be the default model in Google's Gemini app and image generation in the video editing tool, Flow, across 141 countries via Google Lens and in AI Mode.
  • Usage: The model supports high-res image creation suited for storytelling with vibrant lighting, rich textures, and sharper detail. The model incorporates a SynthID watermark for AI-generated images, which ensures interoperability with C2PA Content Credentials, an industry standard.
  • Availability: Available in preview for developers through several Google APIs, including Gemini API, Gemini CLI, and Vertex API.
  • Background: Initially released in August 2025, Nano Banana has been utilized extensively across various regions, including a significant user base in India.