The use case guideline states that:
“We sometimes approve summarizers with a maximum of a paragraph input (150 tokens) and implementation of the OpenAI Content Filter to exclude possibility of ‘Unsafe’ (CF=2) content”
Does the limitation of paragraph input (150 tokens) means the total input from a user or the input of every API call?
Thank you in advance!
Thank you @m-a.schenk!.
If I understand you correctly, taking the user’s input of 600 tokens paragraph and make 4 API calls to summarize the paragraph could be an approvable use case.
Thank you for the detailed explanation! I don’t have a concrete use case yet, and I am trying to get a sense of what kind of use cases could be approved. My first thought is building a quiz generator based on articles that users provide. Talking to OpenAI sounds like a great next step.