scy 🔜 WHY<p><a href="https://chaos.social/tags/Docling" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Docling</span></a> sounds pretty interesting on their website (<a href="https://docling-project.github.io/docling/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">docling-project.github.io/docl</span><span class="invisible">ing/</span></a>), but after having played around with it for a bit, I found the JSON/Markdown/HTML results pretty disappointing.</p><p>OCR was mediocre to bad, table/heading/list recognition too. It didn't even add line breaks between the lines in the address part of a letter.</p><p>But I'm using the defaults. Any suggestions on, like, different models or engines and stuff?</p>