GPT Image Hub / Canvas

AI Design Agent & Infinite Canvas

Generate, edit and arrange AI images on one infinite canvas. Add an image, type a prompt, choose GPT Image 2, Nano Banana or Seedream, and keep every version side by side. No nodes, no setup; sign in and your canvases sync to your account.

• 14 image models, including GPT Image 2, Nano Banana Pro and Seedream 5.0, plus Seedance video
• Use earlier results as references for the next image
• Crop, mask-edit, split and upscale without leaving the canvas
• Start from 31,000+ prompt templates with "Open in Canvas"

FAQ

What is an AI design agent?

An AI design agent turns plain-language instructions into finished visuals. On GPT Image Hub the canvas is the workspace: each image has its own prompt box, you pick the model, and results stay on the board to compare, edit or reuse as references. There is no chat assistant yet; you stay in control of every step.

Do I need nodes like ComfyUI?

No. You never wire nodes together. Add an image to the canvas, write or paste a prompt, choose a model and generate. Connecting an earlier image as a reference is optional.

Which models are supported?

GPT Image 2 (with Saver, 2.5 Flare and 2.5 Sunburst), Nano Banana, Nano Banana 2, 2.1 and Pro, Seedream 4.5, 5.0 and 5.0 Pro, and Z-Image Turbo. Video nodes run Seedance 2.0 Mini, 2.0 Fast, 2.0 and 2.5, from text, a first and last frame, or reference images, videos and audio; the Video page has the same models.

Compare AI canvas tools

Lovart alternative · Krea alternative · FLORA alternative · Best AI canvas tools · Prompt library