Multimodal AI

Image Inpainting API

Repair or regenerate a local image region using a mask and prompt.

post/api/skill-api/v1/skills/image-inpainting/run

Request

  • Base URL: https://api.nexscope.ai
  • URL: https://api.nexscope.ai/api/skill-api/v1/skills/image-inpainting/run
  • Auth: HTTP bearer

Headers

X-Nexscope-Http-Status'legacy' | 'semantic'

Use semantic for HTTP error statuses on run and task-result requests. Without this header v1 retains legacy status behavior. The JSON envelope is unchanged; code 0 means success or acceptance, not asynchronous completion.

Request body

imageUrlstring required

Public URL of the base image. The image must be between 300 and 4096 pixels in both width and height, and its width-to-height ratio must be between 0.4 and 2.5.

promptstring required

Prompt describing how to redraw the masked area.

maskUrlstring required

Public URL of the mask image.

outputNuminteger

Number of images to generate, from 1 to 4. Defaults to 1.

Example request

{
  "imageUrl": "https://example.com/product.jpg",
  "prompt": "remove scratches and keep the product realistic",
  "maskUrl": "https://example.com/mask.png",
  "outputNum": 1
}

Response

Successful response.

codeinteger required

0 means accepted or successful; nonzero is a platform error.

msgstring nullable required
tsstring required

Epoch milliseconds.

timestring required

Server local time: yyyy-MM-dd HH:mm:ss.

coststring required

Elapsed milliseconds, never credits; -1 outside web requests.

traceIdstring nullable required

Changes

No recorded changes to this operation across all 1 revision of this API.