Cat Vton

48 credits/call
ae_cat_vton

ae_cat_vton — Virtual try-on Create a virtual try-on candidate with WaveSpeed Bria Virtual Try-On using a person photo and one to three separate garment or accessory reference images. Supply images as public URLs or Base64-encoded images; the person photo supports JPEG, JPG, PNG, and WEBP. Optional settings include text instructions, output aspect ratio, PNG or JPEG output (PNG by default), and a random seed. If no aspect ratio is specified, the person photo's original aspect ratio is preserved. Processing is asynchronous. Review the result for garment details and apparent fit; the image does not guarantee real-world sizing or fit.

Image
48 credits/call

Pricing

48 credits/call
ConditionCharge
Each call48 credits/call

The fixed fee is 48 credits per call.

The number of garment or accessory reference images, text instructions, output aspect ratio, output format, and random seed do not affect the fixed fee per call.

Parameters

Optional parameters use the platform defaults when omitted.

ParameterTypeRequiredDescriptionDefault
request_bodyobjectYesNo parameter description.—
request_body.seedintegerNoRandom seed for generation. Omit to use a random seed.—
request_body.imagestringYesThe person photo to edit. Public image URL or Base64-encoded image. Supported formats: JPEG, JPG, PNG, and WEBP.—
request_body.promptstringNoOptional instructions to guide the edit. Omit to use the model's built-in instructions.—
request_body.aspect_ratioenum (1:1 | 2:3 | 3:2 | 3:4 | 4:3 | 4:5 | 5:4 | 9:16 | 16:9)NoOptional output aspect ratio. Omit to preserve the person photo's original aspect ratio.—
request_body.output_formatenum (png | jpeg)NoOutput image format. Default: png.png
request_body.reference_imagesarrayYesOne to three garment or accessory reference images. Each entry is a public image URL or Base64-encoded image. These are separate from the person photo in image.—

Invocation example

Copy the request and explore its response example.

POST/agent-api/v1/tool/executeX-Api-Key

Submit first, then query with the returned task_id until the task completes.

Request coderequest.sh
curl --request POST \
  --url 'https://agentearth.ai/agent-api/v1/tool/execute?tool_name=ae_cat_vton' \
  --header "X-Api-Key: $AGENTEARTH_API_KEY" \
  --header 'Content-Type: application/json' \
  --data-raw '{
  "params": {
    "request_body": {
      "aspect_ratio": "16:9",
      "image": "https://cardigan-ref-assets.app.workbuddy.host/person.jpg?sharecode=R6V3NRQMN3",
      "output_format": "png",
      "prompt": "Virtual try-on: replace ONLY her caramel-brown cardigan with the exact cardigan in the garment reference. Keep her pose, hands, hair, face crop, car interior, sunlight, skirt and background unchanged. The cardigan has a wide nearly horizontal neckline with ONE long narrow center-front V slit (left panel diagonal edge, right panel straight vertical edge), EXACTLY THREE small dark round buttons in a centered vertical row placed in the lower half (one at the bottom of the V slit, one on the seam above the wide ribbed waist band, one on the small pointed hem tab), a wide ribbed waist band and a small V-shaped notch at the center bottom hem. No wide open V-neck, no round crew collar, no extra buttons on the upper chest. Photorealistic.",
      "reference_images": [
        "https://cardigan-ref-assets.app.workbuddy.host/garment.jpg?sharecode=R6V3NRQMN3"
      ]
    }
  }
}'
Response exampleresponse.json
JSONPartial content
{
  "description": "task accepted; poll the result via query_tool with this task_id",
  "query_tool": "ae_get_result",
  "status": "created",
  "task_id": "b9ba9c2f-68b3-4f8d-b369-a4f7b6164c27"
}

This example shows part of the response. Copied content is valid JSON.