Notifications
Clear all
Topic starter
Hello,
I'm trying to follow this community guide and have a question: How to PDF into openAI (Solution!)
Does the assistant need to be created based on the vector store, or can I message a different assistant? I'm wondering where I should input the specific guidelines and instructions for this agent, and how to specify the information I want to extract from the document.
Please provide guidance.
Posted : 22/07/2024 9:12 pm
You configure the assistant within OpenAI, utilizing the playground for testing. Subsequently, from the callin.io module, you simply invoke the assistant and provide it with the necessary information.
For instance, within the assistant, you might specify: "I want you to extract the top half of this PDF."
In the callin.io module, you can then pass the PDF directly without any additional instructions, as the assistant has already been pre-configured for this task.
In essence, all your instructions should be established within the OpenAI assistant.
Posted : 24/07/2024 8:47 am