[gvv]

I've had most success using PDFMinerLoader

(https://api.python.langchain.com/en/latest/document_loaders/...)

It deals pretty well with PDF containing a lot of images.