Behrouz Currently, the prompt input does not accept a document as input. It uses text as input to tokenize and pass to the API, so if you are planning to use a PDF document you need to extract the text from the document which needs to be passed as input to prompt and then use the API. With accurate prompt engineering you should be able to efficiently get the result from your model and with fine tuning the model it should further improve the response.
To summarize, PDF file cannot be passed as direct input to the prompt as it is currently not supported.
If this answers your query, do click Accept Answer
and Yes
for was this answer helpful. And, if you have any further query do let us know.