The system uses OCR (Optical Character Recognition) powered by AI to extract nutrition details from nutritional facts label image. The results with nutritional fields like total fat, protein, etc. are structurized and presented as one ingredient in the JSON response.
The processing of nutrition facts label analysis is synchronous. No webhooks will be sent.
precise (default) — uses advanced AI models for the highest accuracy and detail with good processing timefast — uses optimized AI models for good accuracy and detail with the fastest processing timeA predefined list of nutritional fields is extracted. See Nutritional Fields for supported macronutrients and micronutrients.
For more documentation, including implementation examples, processing workflows, and integration guides, see Implementation Guide.
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
OK
Detected ingredient name
"beef and broccoli stir-fry"
Serving size in metric units
x >= 0.01120
Metric unit (g for solids, ml for liquids)
g, mg, mcg, ml, kcal "g"
Reason for processing failure
"Unable to identify label"
Nutritional values in the given serving size
Processing status
completed, failed "completed"