endpoint

OCR

POST /api/v1/ocr

Recognize text in scanned PDFs. 20 languages. Optionally produce a searchable-PDF output.

credits: 5returns: application/json — or application/pdf when searchablePdf=true

OCR is the single most expensive op — price reflects real compute.

Parameters

Name	Type	Required	Description
file	multipart file	required	Source PDF.
languages	string (query)	optional	Comma-separated ISO codes.default: eng
dpi	integer (query)	optional	Rasterization DPI before OCR.default: 200
positions	boolean (query)	optional	Return per-word bounding boxes.
searchablePdf	boolean (query)	optional	Output a text-layer-embedded PDF.

curl

curl -X POST "https://api.snappdf.au/api/v1/ocr?languages=eng,spa&dpi=300" \
  -H "Authorization: Bearer $SNAPPDF_API_KEY" \
  -F "file=@scan.pdf"

JavaScript

const r = await snap.pdf.ocr({ file: bytes, languages: ['eng', 'spa'], dpi: 300 });

Python

r = snap.pdf.ocr(file=bytes, languages=["eng", "spa"], dpi=300)

PHP

$r = $snap->pdf->ocr(file: $bytes, languages: ['eng', 'spa'], dpi: 300);

Ruby

r = snap.pdf.ocr(file: bytes, languages: ['eng', 'spa'], dpi: 300)

r, _ := client.OCR(ctx, &snappdf.OcrInput{File: bytes, Languages: []string{"eng", "spa"}, DPI: 300})