Question 1

What is OpenOCR?

Accepted Answer

OpenOCR is a unified API that lets you access multiple OCR engines (Tesseract, EasyOCR, GPT-4o, Gemini, and more) through a single REST endpoint. Send one request and choose any engine — no need to manage separate provider integrations.

Question 2

How do I authenticate with the API?

Accepted Answer

Include your API key as a Bearer token in the Authorization header: 'Authorization: Bearer sk-ocr-...'. Generate keys from your dashboard after signing up.

Question 3

Which OCR engines are supported?

Accepted Answer

OpenOCR supports 20+ engines including Tesseract, EasyOCR, docTR, OCRmyPDF, RapidOCR, OpenAI GPT-4o, Google Gemini, Anthropic Claude, and more. Each engine has different strengths for handwriting, tables, CJK text, and document types.

Question 4

Does OpenOCR support PDF files?

Accepted Answer

Yes. Send a PDF via the /ocr endpoint and OpenOCR will extract pages and run them through your chosen engine. Multi-page PDFs are supported with per-page billing.

Question 5

Are there SDKs available?

Accepted Answer

Official Python and JavaScript/TypeScript SDKs are coming soon. Today, call the REST API directly with cURL, fetch, Python requests, or any other HTTP client.

Question 6

What image formats are accepted?

Accepted Answer

OpenOCR accepts PNG, JPEG, WebP, TIFF, BMP, and PDF files. Send them as base64 or as a URL reference; multipart file upload is not supported.

Documentation

Documentation