Comment by rullopat

Congrats for the achievement!

I'm curious if this model would be useful for OCR or at least text documents parsing (PDF, DOCX, etc.), what do you think?