Virtual Try-On

Submit a garment image and a model image. A successful request returns HTTP 202 with the task ID.

curl -X POST \
  "https://www.fashiondiffusion.ai/api/public/v1/virtual-try-on/tasks" \
  -H "Authorization: Bearer fd_live_xxx" \
  -H "Content-Type: application/json" \
  --data-raw '{
    "garmentImageUrl": "https://cdn.example.com/garment.jpg",
    "modelImageUrl": "https://cdn.example.com/model.jpg",
    "imageCount": 1,
    "resolution": "2K",
    "ratio": "auto",
    "mode": "FAST",
    "model": "standard",
    "prompt": "Keep the garment shape and fabric texture"
  }'

# Use Base64 instead of URL fields; each image accepts exactly one input form
"garmentImageBase64": "data:image/jpeg;base64,/9j/4AAQSk...",
"modelImageBase64": "data:image/png;base64,iVBORw0KGgo..."

Each image may be a public HTTPS URL or a Base64 data URL. JPEG, PNG, and WebP images must be no larger than 20 MB; the API stores inputs before creating the task.

Each accepted request creates and charges a new task. This version has no idempotency protection, so do not automatically retry a timed-out submission unless duplicate charges are acceptable.