Find answers from the community

Updated 4 months ago

Pa4de

is there a wat to parse a complex pdf extracting tables, images and text, maybe even llm (gpt4o or maybe a local multimodel one)?
maybe while reading files
@Logan M

10 comments

LLogan M

I mean, this is what llama parse does.

If you want, you can send a pdf page by page to a multi modal llm and prompt it to extract too

BBhavya Giri

can i use my deployed locally deployed model?

LLogan M

Yea why not. Just need to send it image of each page and prompt it to extract

BBhavya Giri

still i need to pay for llamaparse, see i am working 1 TB of data
the cost is gonna sky rocket

BBhavya Giri

pdfs are like these, i have 4 A100 to run some opensource llm

BBhavya Giri

any suggestion on approch i should follow

BBhavya Giri

@Logan M

LLogan M

I meant if you have a local multimodal llm running, you could use that instead. Just might take a while

LLogan M

Anything is going to be expensive with 1TB of data my guy lol

BBhavya Giri

do you have any suggested blog post for mulitmodel or something similar to this?

Add a reply