Skip to content
◈

Receipt Scanner

FinanceFree · Private

Scan receipts and extract merchant, date, total, and category entirely in your browser. On-device OCR and AI — your receipts are never uploaded anywhere.

Nothing is uploaded — your receipt photo never leaves this device. No receipt photo, statement, or transaction ever leaves this browser tab — OCR, categorization, and summaries all run locally via WebAssembly. This tool never connects to a bank account or moves money.

Receipt Photo

Upload a photo of a receipt, or take one with your camera.

On-device OCR + AI · nothing uploaded

How the receipt scanner works

  1. Choose or photograph a receipt.
  2. Click Scan Receipt — on-device text recognition reads the text, then a small local AI model and pattern matching pull out the merchant, date, total, and category.
  3. Review the fields — OCR isn't perfect, so correct anything before saving.
  4. Save the receipt, then export everything as CSV or JSON whenever you're ready.

Why some fields are AI and some are pattern matching

The total amount and date are extracted with plain pattern matching, not the AI model — small on-device models are unreliable at reproducing exact numbers, so a currency/date pattern search is both faster and more accurate. The merchant name and category, which need actual language understanding, go through the local LLM instead.

Where is my data stored?

Saved receipts live in your browser's local storage only — on this device, in this browser. Nothing is synced or uploaded. Clearing your browser data will also clear your saved receipts, so export anything you want to keep.

Frequently Asked Questions

How does the receipt scanner work?
It runs an on-device text-recognition engine to read the text on your receipt photo, then a small local AI model and pattern matching extract the merchant, date, total, and category from that text. Everything happens inside your browser tab.
Is my receipt photo uploaded anywhere?
No. The image never leaves your device — text recognition and field extraction both run locally in your browser. Nothing is sent to a server at any point.
What if the OCR misreads something?
Every extracted field (merchant, date, total, category) is shown in an editable form before you save it, so you can correct any OCR mistakes — receipt photos vary a lot in quality and this is common.
Can I export my scanned receipts?
Yes — once you have saved receipts, use the Export button to download them as a CSV (for spreadsheets) or JSON file, generated entirely client-side.