Hi everyone,
I’m experiencing issues with processing PDF files using the OpenAI Assistant API. When trying to get ChatGPT to analyze PDF content, I consistently receive the error:
“It seems there is an issue accessing the uploaded file. Could you please try uploading the document again?”
My current setup:
- Watch files in Google Drive (monitoring PDF files)
- Download the latest PDF file from the folder
- Upload to OpenAI and process with Assistant API
I’ve followed this tutorial from June 28th: How to PDF into openAI (Solution!)
About setting up file processing with vector stores, including:
- Using the OpenAI upload module
- Creating and using a vector store
- Enabling file search
- Setting up the Assistant with proper file references
However, I’m noticing that my PDF file output is coming through as binary data instead of readable text.
Can anyone help me with this?
Any guidance would be greatly appreciated!