Dev: OCR Playground

Try google (Vision API), pytesseract, openai (vision), or mistral (OCR) on PDFs/images.

Settings

BackendPDF DPIMax Pages (0 = no limit)Max Chars (0 = no limit)Google language hints (optional)Tesseract lang (optional)

Run OCR

From local path (server-side)

Allowed root: DEV_OCR_ALLOWED_ROOT (default /Users/adnanehamid/ttc/customers)

Or upload a file

Result

backend: -

method: -

chars: -

durationMs: -