Where DocFila runs OCR on your phone, where it can use a cloud model instead, and the trade-offs to know.
Cloud OCR sends the entire document to a server. On-device OCR keeps the page on the phone. For medical records, IDs, legal contracts, and anything sensitive — the difference matters.
DocFila runs OCR on the device, using Apple Vision on iOS 16 and later and Google's ML Kit on Android, for Latin, Chinese, Japanese, Korean and Devanagari text. Arabic, Cyrillic and Greek have no on-device model: when the on-device pass finds nothing, a signed-in member can send the page to DocFila's AI service. Guest sessions never do. PDF parsing and invoice extraction can also send the document itself when on-device extraction finds too little text. See /ai-data-handling for the full list.
Many cloud OCR services bill per page. On-device OCR has no per-page fee.
On-device OCR reads typed text in good light well. Poor handwriting, very degraded scans and scripts without an on-device model are where an AI model can do better. DocFila has not benchmarked on-device against cloud OCR, so we do not claim the two match.
Keep going with the document workflow that matches this page. Open the tool you need, then finish the file in DocFila.
Scan, sign, edit, translate, and store your documents in one app. Free to download, with optional paid features. No ads.
Or open the web app in any browser.