I have a few API calls that run sequentially. Most of the time, they execute without any issues. However, in some cases — especially under concurrent requests — a few calls fail due to token limit restrictions, causing the entire sequence to break.
I’m looking to find out if OpenAI provides any queue-based mechanism for API calls, so that when a call fails, it can be retried from the point of failure rather than restarting the entire sequence.
Thanks
Ram