News for hackers. Stuff that matters.
Re:Show HN: Zerox – document OCR with GPT-mini by themanmaran on Wednesday July 24, 2024 @02:14AM (#41054072)
I think that I'll add an optional configuration for HTML vs Markdown. Which at the end of the day will just prompt the model differently.

I've not seen a meaningful difference between either, except when it comes to tables. It seems like HTML tends to outperform markdown tables, especially when you have a lot of complexity (i.e. tables within tables, lots of subheaders).

There are no comments on this comment yet.