API Key for API endpoints
Text prompt describing the desired output or edit. Supports Chinese and English.
1 - 2000Number of results to generate (1-4)
1 <= x <= 4Optional reference images for edit mode (1-4 images). Each image must be at least 384x384 pixels. When provided, the model runs image edit.
4Content to avoid in the generated image. Max 500 characters.
500Aspect ratio used to derive output size.
1:1, 3:4, 9:16, 4:3, 16:9 Output image size preset. Advanced override.
square_hd, square, portrait_4_3, portrait_16_9, portrait_9_16, landscape_4_3, landscape_16_9 Enable prompt expansion. Only used in edit mode.
Random seed for reproducibility (0-2147483647).
0 <= x <= 2147483647Enable content moderation for input and output.
Number of images to generate. Text mode supports up to 5; edit mode supports up to 4.
1 <= x <= 5Maximum number of images to generate (1-5). Used by text mode.
1 <= x <= 5