Image-01 Image to Image
Image-to-image capability for the Image-01 model.
Exposed fields: model, prompt, subject_reference, aspect_ratio, width, height, response_format, seed, n, prompt_optimizer. subject_reference is expanded as a JSON request in this document.
Authorizations
Pass Authorization: Bearer <token> in the request header.
Body
Model name for this capability.
image-01 Image editing prompt.
1500Subject reference information, used to enforce character or object consistency in the generated result.
1Target aspect ratio.
1:1, 16:9, 4:3, 3:2, 2:3, 3:4, 9:16, 21:9 Generated image width (pixels). Only effective when model is image-01. Note: width and height must be set together, range [512, 2048], and must be multiples of 8. If set together with aspect_ratio, aspect_ratio takes priority.
512 <= x <= 2048Must be a multiple of 8Generated image height (pixels). Only effective when model is image-01. Note: width and height must be set together, range [512, 2048], and must be multiples of 8. If set together with aspect_ratio, aspect_ratio takes priority.
512 <= x <= 2048Must be a multiple of 8Response format. Use url to get a temporary download link, or base64 to get Base64-encoded content.
url, base64 Random seed.
Number of images to generate.
1 <= x <= 9Whether to enable prompt optimization.