Home / APIs / OCR API

OCR API for Images & PDFs

Run OCR on images via POST /v1/ocr (sync JSON text) and on PDFs via async /v1/pdf/ocr or make-searchable flows. Image OCR uses Tesseract with a language parameter; PDF OCR supports eng/ara in the PDF pipeline.

Get API keyTry Image OCRDocumentation

Features

What this API actually supports in production today.

  • Image OCR: multipart file + lang (default eng)
  • PDF OCR async jobs
  • Make-searchable PDF pipeline
  • Related document extract / invoice scan tools
  • Same API key as other FatBaby products

Common use cases

  • Digitize scans and receipts
  • Searchable archives from scanned PDFs
  • Pre-processing for document workflows

API endpoints

Base URL https://thefatbaby.com/api. Authenticate with Authorization: Bearer fb_live_YOUR_KEY.

MethodPathSummaryCredits
POST/v1/ocrOCR an image to text3 credits default
POST/v1/pdf/ocrOCR a PDF (async)Per-page OCR billing
POST/v1/pdf/make-searchableMake PDF searchableOCR-related billing

Examples

Primary endpoint /v1/ocr · Image OCR default 3 credits; PDF OCR billed per page (runtime rates apply).

cURL

curl -X POST https://thefatbaby.com/api/v1/ocr \
  -H "Authorization: Bearer fb_live_YOUR_KEY" \
  -F "file=@scan.png" -F "lang=eng"

JavaScript

const form = new FormData();
form.append('file', imageFile);
form.append('lang', 'eng');
const res = await fetch('https://thefatbaby.com/api/v1/ocr', {
  method: 'POST',
  headers: { Authorization: 'Bearer fb_live_YOUR_KEY' },
  body: form
});
const data = await res.json();

Python

import requests
r = requests.post(
  'https://thefatbaby.com/api/v1/ocr',
  headers={'Authorization': 'Bearer fb_live_YOUR_KEY'},
  files={'file': open('scan.png','rb')},
  data={'lang': 'eng'},
).json()
print(r.get('text') or r)

Request

multipart/form-data
file=@scan.png
lang=eng

Response

{
  "text": "...",
  "credits_used": 3,
  "request_id": "req_..."
}

Pricing & credits

Image OCR ~3 credits; PDF OCR uses per-page pricing.

View pricingStart free

Common errors

  • 400 — Unsupported image or missing file
  • 402 — insufficient_credits

FAQ

Is accuracy guaranteed?

OCR quality depends on scan quality and language. We do not claim 100% accuracy.

Image vs PDF?

Use /v1/ocr for images. Use /v1/pdf/ocr for multi-page PDFs (async job).

Try without code

Use the matching online tool, then automate the same workflow with this API.

Try Image OCRGet API key