Cool experiment. I've been mostly using Claude with the various Chat apps - currently with LibreChat - and it does a pretty good job!

Quite curious how this compares to docling - https://github.com/DS4SD/docling

docling uses an LLM IIRC, so that's already a difference in approach

In my use, docling has not involved an LLM. There are a few choices for OCR, but I don't think a vision model is one of them.

It's certainly touted as a solution to digest documents into plain text for LLM use, but (unless I just haven't run into that part of it) it does not employe an LLM for its functions.

docling does not use LLMs...

You might like Michael Drogalis' blog - https://substack.com/@michaeldrogalis

He's a software engineer who's been building his business in the open for the last year and is sharing what he learns along the way.

This is so great! Thanks for the reply, going through it rn.