I have a strange one here. I have created a new collection in an existing model and used 6 documents to train the model. The model detected all the text but displays none of of the text. I get the same result when using the Power App PDFViewer in an App.
Any ideas on how to resolve this?