How to generate and edit images
Write the picture you want, or upload a reference; when something falls short, select that area and change only it.
- 01
Pick the model
Nano Banana, GPT-Image-2 and Seedream 5.0 Pro all sit in the same editor, so you can try another one on the same prompt instead of moving to another tool.
- 02
Describe it, or show it
Write what the subject is and how it should look, and add up to five reference photos when words alone will not pin it down.
- 03
Fix one region at a time
Select the part that is wrong and say what it should be instead. The rest of the frame is untouched, so a fix never costs you the parts you already liked.
- 04
Step back through versions
Every generation and every edit is kept as its own version. Compare them, return to any one, and download the original file.
AI Image Generator FAQ
Which models can I use?
Nano Banana, GPT-Image-2 and Seedream 5.0 Pro. You switch between them in the same editor, and the results from each sit together in the version history for comparison.
Can I edit part of an image instead of regenerating it?
Yes, that is the point of the editor. Select a region, describe the change, and only that area is regenerated. The edit is saved as a new version rather than replacing the previous one.
What is an AI image generator?
It turns a written description, or a photo you upload, into a new image. You describe the result instead of drawing or shooting it.
Can I start from my own photo?
Yes. Upload up to five reference images in JPG, PNG, or WebP and they steer the generation.
