Google has officially launched Nano Banana 2 (formally known as Gemini 3.1 Flash Image), a significant update that merges the high-end creative control of the “Pro” models with the efficiency of the “Flash” series.
Announced on February 26, 2026, the model is designed to offer production-level quality at a fraction of the previous generation’s latency.
Key Features & Enhancements
- Lightning Speed & High Fidelity: Built on Gemini 3.1 Flash, it enables rapid image generation and near-instant editing iterations. It supports resolutions ranging from 512px to 4K.
- Search Grounding: Unlike static models, Nano Banana 2 uses real-time web grounding via Google Search to accurately render specific real-world landmarks, products, or current events.
- Precision Text & Translation: It offers advanced in-image text rendering (avoiding the typical “AI gibberish”) and supports direct localization, allowing you to translate text within an image instantly.
- Subject Consistency: The model can maintain the visual identity of up to five characters and 14 objects across a single workflow, making it ideal for storyboarding or multi-scene marketing campaigns.
- Expanded Creative Control: It introduces new aspect ratios (including 1:4, 4:1, 1:8, and 8:1) and provides a “Thinking” mode that allows the model to reason through complex, multi-layered prompts before rendering.
Availability
The rollout is currently underway across the Google ecosystem:
- Gemini App: It is now the default model for “Fast,” “Thinking,” and “Pro” modes.
- Google Search: Integrated into AI Mode and Google Lens across 141 countries.
- Flow & Ads: Powering creative suggestions in Google Ads and serving as the default generator in Flow.
- Developer Access: Available in preview via the Gemini API and Vertex AI.
