Nano Banana 2 in ArtCraft: The Image Model Behind Most of the Workflow

What Nano Banana 2 is

Nano Banana is the codename for Google's Gemini-family image generation and editing line. Version 2 improves editing fidelity, text rendering inside images and multi-reference conditioning - the ability to take several input images and blend their content coherently.

Why it matters inside ArtCraft

ArtCraft is built around editing and control as much as generation - and editing is where Nano Banana historically stands out. Compose a scene on the canvas, then let the model harmonize lighting and style across the elements, or mask a region and re-render just that part.

• Text-to-image: clean generations with unusually good in-image text.
• Image editing: upload a photo, describe the change, keep everything else.
• Multi-reference: blend several input images into one coherent output.
• Character identity transfer: keep a face consistent across poses and scenes.

Strengths

• Best-in-class text-in-image rendering (signage, posters, UI).
• Editing over raw generation - it changes what you ask and preserves the rest.
• Multi-image conditioning fits ArtCraft's canvas workflow naturally.
• Fast enough for tight iteration loops.

Limitations to know

• Stylization range is narrower than art-focused models like Midjourney.
• Aspect-ratio control exists but extreme crops still benefit from canvas composition.
• Provider-metered - generation costs apply like every model in the catalog.

Try it in the browser

Our text-to-image tool runs a Nano Banana-class model with reference-image support, and the image editor uses the same family for upload-and-edit workflows - no install needed.