Contact Card Reader

1. Capture or upload a business card

Snap a photo with your phone's rear camera, or upload an existing image (JPEG / PNG / WebP). By default the photo is uploaded to the server for OCR (better accuracy and ~3× faster than running in the browser); the image is processed in memory and never written to disk or logged. With Use LLM on, the OCR text is then handed to a small local model for structured extraction — the model runs on the same cluster, so nothing leaves the cluster network. Uncheck Server-side OCR below to run everything locally in your browser instead — useful if you'd rather the image not leave your device.

Drag & drop a business card image here

or use the buttons above

Show preprocessed image (what Tesseract sees)

2. Extract & edit contact details

Click Extract to OCR the image and auto-fill detected fields. Tesseract.js can misread stylised cards — review and correct everything before generating the vCard.

Server-side OCR (recommended) Use LLM (better accuracy)

Browser-only preprocessing: Preprocess Auto-crop

First name

Last name

Full name (FN)

Title

Organization

Website

Phone numbers

Email addresses

Address

Notes

Show raw OCR output

— conf 40-69 conf <40

3. Save the vCard

Download the .vcf to import into Contacts / Outlook, or email it via the [email protected] relay. Sending email requires a valid recipient.