Identify Different Level of Headings

[aswincandra]

Hi, I've tried the minerU to extract the PDF document, then I realized, according to the markdown result, you can perfectly detect the heading, but the level of headings are always identified as level-1 heading. Can you guys identify different level of headings? (Heading 1 #, Heading 2 ##, and so forth), e.g. by the font size.

[grappa-ytannous]

Seems like heading classification is still a feature in preview
Only released for the online demo
But we need it for API-use as well