SAM 3 Image
Image editing
Starting price $0.01 / request
Microsoft's Florence-2 Large outlines the part of a photo you describe in words, such as "the dog on the left", and returns its exact shape.
florence-2-large/referring-expression-segmentationQuote the input, start the generation with the quote, then poll the job. Set RAYWAKE_API_KEY first; see Authentication.
# 1. Price the exact input (quotes are valid for up to 10 minutes)
curl -s https://api.raywake.com/v1/quotes \
-H "Authorization: Bearer $RAYWAKE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "florence-2-large/referring-expression-segmentation",
"input": {
"image_url": "https://example.com/input.png",
"text_input": "the dog on the left"
}
}'
# 2. Start the generation with the quote_id from step 1
curl -s https://api.raywake.com/v1/generate \
-H "Authorization: Bearer $RAYWAKE_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $(uuidgen)" \
-d '{"model": "florence-2-large/referring-expression-segmentation", "quote_id": "QUOTE_ID", "input": {"image_url":"https://example.com/input.png","text_input":"the dog on the left"}}'
# 3. Poll until the status is terminal
curl -s https://api.raywake.com/v1/jobs/JOB_ID \
-H "Authorization: Bearer $RAYWAKE_API_KEY"| Name | Type | Description |
|---|---|---|
image_urlrequired | string (image URL) | The URL of the image to be processed. |
text_inputrequired | string | Text input for the task |
A succeeded job lists its results under outputs, stored on our side. See Jobs for every field.
"outputs": [
{ "type": "image", "url": "https://…", "mime_type": "image/png", "thumbnail_url": "https://…" }
]The exact charge for your input comes from a quote. How charges settle
Fields marked * are required. Review the price before generating.
Upload an image and write a short phrase that points to one thing in it. You get back the image with that region marked, plus its outline as a list of polygon points you can use as a selection or mask.
Compare returned polygon boundaries with the intended object. Inspect thin parts and adjacent subjects before using the selection.