135
Views
0
Saves
About OCRZero
OCRZero reads scanned pages, photos and PDFs with the Xiaomi-OCR-0 model and gives you clean output you can actually reuse: Markdown that keeps headings, tables and formulas, or JSON fields pulled from invoices and forms. Try it free in the browser for one-off files, or call the OCR API to batch-process documents for bookkeeping, archives or data pipelines.
Updated 10/31/2026
Key Features
- Markdown, tables, and LaTeX formulas from document images and PDFs
- Invoice and form field extraction to structured JSON
- API access for batch processing and document Q&A
- 20 free pages for new accounts
- 1 credit per page with automatic refunds on failure
- $5 per 1,000 pages for paid credits
- Self-hostable with vLLM for private deployments
Pricing Structure
- Free tier — $0 — 20 free pages for new accounts; browser access to all output formats
- Paid credits — $5 per 1,000 pages — Credits never expire; refunded automatically on failed pages
Why People Use It
OCRZero is built for teams that need to turn scanned documents into reusable data without manual transcription. The combination of free online access and cheap API credits makes it practical for both one-off PDFs and high-volume archives. Open-source model licensing also appeals to teams that want to self-host for privacy or cost control.
See OCRZero in Action
Tags
0
selected