API reference/POST /smart-crop

Auto-crop to subject bounds

Tight bounding box around the subject. Optional padding. Useful for thumbnails, e-commerce tiles, and uniform aspect ratios across product shots.

POSThttps://useknockout--api.modal.run/smart-crop

Parameters

Send as multipart/form-data unless noted otherwise.

filerequiredfile–Image to process. JPG, PNG, WebP, HEIC. Up to 10 MB and 4096×4096.
paddingint24Pixels of padding around the subject's tight bounding box.
transparentbooleantruetrue returns the cropped cutout on a transparent background. false returns the cropped region of the original image, background kept.
detectstringstandardSubject detection mode: standard (default) or high_recall. high_recall runs a second model pass on a contrast-boosted copy and unions the two masks, recovering product regions the single pass misses on low-contrast shots (a pale product on a pale surface). It only ever adds to the mask; it never removes included background (that's decontaminate). Runs double inference, so latency roughly doubles and it requires a paid plan: free-tier requests get a 402.
decontaminatebooleanfalseRe-classify pixels near the mask edge using per-image color models of subject vs background, removing background the model kept: base-paper strips under products, lightbox rig fragments at the boundary. Trade-off: it helps least on a white product on a white background (no color evidence to work with), and it can trim the dark underside edge of stacked or thick products, which is partly real product. Available on every tier.
enginestringstandardWhich cutout engine runs: standard (default), auto, or product-v1. Use auto for product photography: it runs the standard engine, then switches to product-v1 for flat products (packaging film, paper, sheet goods, lightbox flat-lays) and rescues dark subjects on flat backdrops where the default cut drops them. auto only costs extra time when it escalates, typically 45 to 70 seconds instead of 15 to 30. product-v1 forces the product engine on every image; avoid it for plants, hair, and translucent or pale organic subjects. auto and product-v1 need a paid plan on signed-in accounts (free accounts get a 402). The X-Knockout-Engine response header says which engine actually ran.
formatstringpngOutput format: png or webp; jpg also accepted when transparent=false.

Request

curl -X POST "https://useknockout--api.modal.run/smart-crop" \
  -H "Authorization: Bearer $TOKEN" \
  -F "file=@cat.jpg" \
  -F "padding=16" -F "transparent=true" \
  -o out.png

Response

HTTP/1.1 200 OK
content-type: image/png
content-length: 254312
x-knockout-latency: 184
x-knockout-model: BiRefNet
x-knockout-engine: standard
x-ratelimit-limit: 60
x-ratelimit-remaining: 59

Errors

401unauthorized–Missing or invalid token.
402payment_required–Free tier exhausted. Add a card to continue.
413payload_too_large–Image exceeds 10 MB or 4096×4096.
422no_subject_detected–Foreground could not be isolated from background.
429rate_limit_exceeded–Slow down. Retry-After header tells you when.
500internal_error–Something broke on our side. Include request_id when reporting.
Every error response also includes a request_id in the JSON body. Quote it when reporting issues.

Notes

  • The crop is the subject's tight bounding box plus padding, so output size follows the subject. For a fixed aspect ratio on a canvas, use /studio-shot.
  • decontaminate=true also tightens the crop, since background strips no longer inflate the box.
← POST /outlinePOST /silhouette →