Nano Banana Pro brings Gemini-3 image generation
Image2026-06-12
Source: blog.google
Google DeepMind's Nano Banana Pro sets a new bar for in-image text and precise editing, intensifying the race with GPT Image and FLUX.
Google's latest image model leans into the two things diffusion has historically been worst at: rendering legible text inside an image, and making a small, controlled edit without redrawing everything else. Nano Banana Pro does both well enough that it changes what the tool is good for — posters, packaging mock-ups and UI screenshots, not just illustration.
It lands in a crowded field. See how the hosted services compare with the open models you can run yourself on the image generation page — FLUX, Stable Diffusion and Z-Image all run locally on a 12–24 GB card through ComfyUI.