User ID: Loading...
Model
Pick the style engine. The card below explains what the selected model is best at.
Collection
Group images by project or client so repeated runs stay organized.
Image Prompt
Describe the image you want. You can type in any language — the system will auto-translate to English.
Negative Prompt
What you don't want in the image, e.g. blurry, low quality, watermark, deformed hands
Input Image + Prompt Edit
Add a source image, then use the prompt to change it. Paint a mask only when you want to edit a specific area.
Aspect Ratio / Image Size
Select an aspect ratio or enter custom width × height in pixels.
Allowed range: width 256-1536, height 256-1536, step 64px.
Steps
Number of inference steps. This model supports 1-80.
Guidance Scale
How closely the model follows the prompt. This model supports 0-20.
Seed
Controls randomness. Same seed produces similar results. Leave empty for a random seed.
No images yet
Enter a prompt and click Generate to create your first image.