idle
⚠️ Please don't upload confidential or private documents. This is a research demo and uploaded images are collected to train this OCR model that will be open-sourced in the future. This project runs on a lot of compute — if you find it useful, a small donation helps keep it running. 🙏 Buy me a coffee / donate · 📚 want to donate a dataset? Contact mrrtmob@gmail.com.
click, drop, or paste an image
Ctrl+V anywhere · Ctrl+Enter to run
examples

Result

step committed canvas elapsed