Choose an approach
Choose how to read your documents and extract useful information.
Extract from text or from the PDF?Extract from the text column when it holds what you need. Send the PDF itself when the answer depends on how the page looks.
Which model should extract your fields?Start with a cheap, fast model and test it on a few rows. Move to a bigger one only if fields come out wrong, then check the same rows again.
Which OCR engine should you use?Use the PDF's own text if it's good. If it isn't, start with a local engine and try a hosted one on the pages it gets wrong.