olmOCR

github.com ↗

- olmOCR is a toolkit for training language models to work with PDF documents in the wild.