Nano Banana 2: Google rebuilds its image model around consistency and text
In one sentence Google releases Nano Banana 2 (Gemini 3.1 Flash Image): consistency for up to 5 characters and 14 objects in one workflow, legible and translatable in-image text, output from 512px to 4K. It replaces Nano Banana Pro in the Gemini app, becomes the default in Flow, and reaches AI Mode and Lens in 141 new countries and territories.
Nano Banana is the name Google uses for its image generation and editing models. On February 26, 2026 it released Nano Banana 2, whose technical name is Gemini 3.1 Flash Image.
What Google promises:
- keeping up to 5 characters and up to 14 objects consistent within the same workflow;
- more legible text inside images, useful for mockups and cards, with the option to translate and localize it within the image;
- closer adherence to complex instructions;
- images from 512 pixels up to 4K, in multiple aspect ratios.
In the Gemini app it replaces Nano Banana Pro in the Fast, Thinking, and Pro models (Google AI Pro and Ultra subscribers can still regenerate an image with Nano Banana Pro). In Flow it becomes the default image model for everyone, at zero credits. In Search it arrives in AI Mode and Lens, with 141 new countries and territories and eight more languages. Images carry SynthID and C2PA content credentials.
Companies
Google, Google DeepMind
Tools
Nano Banana 2, Gemini 3.1 Flash Image
Tags
Sources