OCR (documents)

Extract text from images and PDFs (including multi-page PDFs) and get markdown back.

POST https://api.norlen.io/v1/ocr
Authorization: Bearer YOUR_API_KEY

Request

multipart/form-data — send the file to process.

Field	Type	Required	Description
`file`	file	yes	Image (`png`, `jpg`, `webp`, `bmp`, `tiff`) or `pdf` — up to 20MB
`clean`	string	no	`true` strips layout/detection markup; defaults to `false`

curl https://api.norlen.io/v1/ocr \
  -H "Authorization: Bearer $NORLEN_API_KEY" \
  -F file="@document.pdf" \
  -F clean="true"

import requests

resp = requests.post(
    "https://api.norlen.io/v1/ocr",
    headers={"Authorization": f"Bearer {NORLEN_API_KEY}"},
    files={"file": open("document.pdf", "rb")},
    data={"clean": "true"},
)
print(resp.json()["text"])

import fs from "fs";

const fd = new FormData();
fd.append("file", new Blob([fs.readFileSync("document.pdf")]), "document.pdf");
fd.append("clean", "true");

const resp = await fetch("https://api.norlen.io/v1/ocr", {
  method: "POST",
  headers: { Authorization: `Bearer ${process.env.NORLEN_API_KEY}` },
  body: fd,
});
const { text } = await resp.json();
console.log(text);

Response

{
  "text": "# Document title\n\nExtracted content as markdown..."
}

Multi-page PDFs are processed in sequence and concatenated into the same text.