Particle.news

Google Launches Nano Banana 2 as Default AI Image Model Across Its Ecosystem

The faster Gemini 3.1 Flash Image system brings Pro‑level fidelity with built‑in provenance labels.

Overview

  • Officially named Gemini 3.1 Flash Image, the model is now the default for image generation in the Gemini app and is rolling into Search’s AI Mode and Lens, Flow, and Google Ads across 141 countries, with Nano Banana Pro still accessible to Google AI Pro and Ultra subscribers.
  • The release targets Pro‑grade results at Flash speeds, supporting outputs from 512px up to 4K with richer textures, sharper details, and rapid iteration.
  • Text rendering and translation are more accurate, instruction‑following is stronger, and the system maintains subject consistency for up to five characters and fidelity for up to 14 objects in a single workflow.
  • All outputs are marked with SynthID and support C2PA Content Credentials, and Google says SynthID verification in Gemini has been used more than 20 million times since November.
  • Developer previews are available via AI Studio, the Gemini API and CLI, Vertex AI, and Antigravity, while separate APK‑insight reporting points to unconfirmed internal testing of Nano Banana features in Google Maps.