Make your PDF searchable — runs entirely in your browser, nothing uploaded.
Every major online OCR service works the same way: you upload your file, their server processes it, and you download the result. Your document passes through their infrastructure — stored, at least briefly, on their systems. For contracts, medical records, tax returns, or confidential business documents, that is often unacceptable.
PDFree works differently. The OCR engine runs inside your browser tab using WebAssembly. Your file is read from your device into browser memory, processed entirely locally, and the result is written back to your device. No server receives your data — not for a millisecond.
PDFree uses Tesseract.js v5 — the industry-standard open-source OCR engine compiled to WebAssembly. When you click Install OCR Engine, your browser downloads the engine (~17 MB) once and caches it permanently. Every subsequent OCR job runs fully offline from that cached engine.
You don't have to take our word for it. Here is how to verify during a real OCR job:
This is verifiable, not a privacy policy promise. The architecture makes upload impossible — there is no code path that sends file bytes to any server.
| Use case | Why local matters |
|---|---|
| Medical records | Patient data must not pass through unapproved third-party systems |
| Legal documents | Attorney-client privilege; contracts and court filings must stay confidential |
| Financial records | Tax returns and bank statements contain personal data that must not be stored externally |
| HR and payroll | Employee contracts and salary documents are subject to data protection regulations |
| Government forms | Passport scans, ID documents, and official filings must not be uploaded to consumer services |
| Corporate IP | Technical drawings, patents, and proprietary specifications should not leave the organization's control |
Once the Tesseract.js engine and your chosen language models are cached, PDFree works with no internet connection at all. This is useful in secure environments that restrict external network access, on flights, or in areas with unreliable connectivity. Open the app, drop your scanned PDF, and process — no network request is made.
PDFree does not receive, store, or transmit protected health information — your files never leave your device. This removes the primary HIPAA compliance burden. However, PDFree is not a formally certified Business Associate and does not sign BAA agreements. For organizations that require a signed BAA, consult your compliance team about whether local browser-based processing satisfies your specific requirements.
PDFree uses Plausible Analytics — a privacy-first, cookieless tool that counts page views and events without tracking individual users or storing personal data. No file names, no file contents, and no personally identifiable information are ever collected. Plausible does not use cookies and is GDPR-compliant by design.
18 languages, all processed on-device: English, French, German, Spanish, Italian, Portuguese, Russian, Uzbek, Dutch, Polish, Turkish, Arabic, Japanese, Chinese Simplified, Chinese Traditional, Korean, Hindi, and Thai. Language models are downloaded once and cached — repeat use requires no network.
Files up to 200 MB are supported with no page limit on desktop browsers. On iOS and iPadOS, processing is capped at 30 pages per run to stay within browser memory. Use Split PDF to process larger documents in sections.