It has an interesting note: `Currently, the time to first token you receive from a background response is higher than what you receive from a synchronous one. We are working to reduce this latency gap in the coming weeks.`
A webhook by nature has a latency anyway, but it seems like the latency is much longer than one would expected such that OpenAI has to add this note???