The Story: Google has rolled out Nano Banana 2, the upgraded version of its viral image model, delivering higher resolution, improved consistency, stronger text rendering, and faster performance — at roughly half the price of its predecessor.

Leaderboard Performance
Nano Banana 2 reportedly beat both NB Pro and GPT Image 1.5 to claim the No.1 position on major text-to-image benchmarks including Artificial Analysis and LM Arena. It also ranked among the top models for editing tasks.

Quality and Scale
The model supports output scaling up to 4K across multiple aspect ratios. Google says scenes can maintain visual consistency across up to five characters and 14 objects, addressing one of the most persistent weaknesses in generative image systems.

Speed and Pricing
At approximately seven cents per image, Nano Banana 2 undercuts its predecessor and competing models by nearly 2×, while delivering speeds comparable to Gemini Flash-class models.

Ecosystem Integration
Nano Banana 2 is now integrated as the default image generator across Gemini and Google’s broader tool ecosystem, with Pro variants still available for paid subscribers.

Why It Matters: The update signals a compression of the traditional tradeoff between quality, speed, and cost in image generation. If frontier-level output can be delivered at flash-level pricing, the barrier to high-volume creative AI workflows continues to fall.