Create task black-forest-labs/flux-fill-pro
Official document: https://replicate.com/black-forest-labs/flux-fill-pro
/replicate/v1/models/black-forest-labs/flux-fill-pro/predictionsBody Parametersapplication/json
Input content
propertiesControl the balance between text prompt compliance and image quality/variety. Higher values bring the output closer to the prompt, but may reduce overall image quality. Lower values provide more creative freedom in the output, but may be less relevant to the prompt. Default value: 60. Minimum value: 1.5, maximum value: 100.
Pictures to be repaired. Can include alpha masks. Must be in jpeg, png, gif or webp format.
A black and white image describing the portion of the image that needs restoration. Black areas will be preserved and white areas will be repaired. Must be the same size as the image. This parameter is optional if you provided an alpha mask in the original image. Must be in jpeg, png, gif or webp format.
Options for quick repair of input images. Masks will be ignored. Default value: "None"
The format of the output image. Default value: "jpg".
Image-generated text prompts.
Automatically modify prompts to generate more ideas. Default value: false
Security tolerance, with 1 being the strictest and 6 being the loosest. Default value: 2.
Number of diffusion steps. Higher values provide finer detail, but also take longer to process. Default value: 50. Minimum value: 15, maximum value: 50.
Responses
Creation time
Whether sensitive data has been removed
Error message
Resource or task ID
Input content
Run logs
Model name / model ID
Output result
Status
List of resource URLs
Version
