#image-models

This paper argues that image generators may be turning into the vision equivalent of large language models. DeepMind says Vision Banana, built on Nano Banana Pro, beats or rivals specialist systems such as Segment Anything and Depth Anything on 2D and 3D tasks after lightweight instruction tuning.

Google AI used X on March 6, 2026 to direct developers to Nano Banana 2, saying the model is available through the Gemini API in Google AI Studio and Vertex AI. Google’s linked post positions Nano Banana 2, or Gemini 3.1 Flash Image, as a high-quality and faster image model designed for real application workloads.