What we're building
An OCR API you can use today, a scanner app for your phone, and a digitization service for larger collections. All of them run on our own models.
Preview live
Runemic API
Send a page image and get its text back in reading order, as Markdown or plain text, with tables kept. Use it when you need document text in your product, search index or LLM pipeline and don't want to run models yourself. The preview runs open models; Rune-1 replaces them when it's ready.
- Today: PNG, JPEG, WebP or GIF in; Markdown or plain text out. PDF input and JSON output come with Rune-1
- Priced per page, with a lower rate for batch jobs later
- We don't use your documents for training
- Sign in with GitHub or Hugging Face and get $5 of free credit
# Preview models today; Rune-1 when it launches curl https://api.runemic.com/v1/ocr \ -H "Authorization: Bearer $RUNEMIC_KEY" \ -F "file=@letter.png" \ -F "model=preview-gemma-4" # → Markdown text, in reading order { "object": "ocr.result", "text": "# Letter\n\nDear …", "usage": { "pages": 1, "cost_usd": 0.002 } }
In development
Runemic Scan
A scanner app for everyday paper. Take a photo of a receipt, bill, letter or contract, and Runemic Scan straightens it, reads it, picks out amounts and due dates, and files it.
- Keeps track of bills and receipts
- Search everything you've scanned
- Reads Arabic, Persian, Hindi and other scripts, not only Latin
- Export to PDF, spreadsheets or accounting software
Taking enquiries
Digitization services
For libraries, archives, publishers, universities, companies and families with boxes of paper. We plan the project, scan on-site or in our studio, run OCR, check the results and deliver searchable files and structured data. See how projects are priced.
Libraries and archives
Books, manuscripts, newspapers and records, with handwritten text recognition for historical hands, delivered in open formats such as PDF/A, ALTO and IIIF.
Businesses
Contracts, invoices and old files, turned into data your systems can search.
Personal archives
Family letters, diaries and documents: scanned, transcribed and kept private.
Research
Scanning hardware
Book and document scanners designed around our models, with cradles, lighting and cameras set up to give the OCR clean images. We'll build them for our own digitization work first, then offer them to libraries that want to scan in-house.
On your own machines
Self-hosted and on-device options, so sensitive documents don't have to leave your building.
For institutions
Procurement, security reviews and long-term support for public bodies and companies.