This error is due to an overloaded system.
Has it happened recently? Does it happen all the time? Does it happen at a specific time of day?
What model do you use? Because OpenAI 3.5 cannot handle large volumes of data and generates this type of failure.
I suggest that you initially update your AI model to a stronger model with the capacity to handle a larger volume of TOKENS.
Then run your flow again to see if the failure persists.
Check the logs for messages about high demand within OpenAI, because it may be a problem with OpenAI’s own server;
I hope I have helped in some way.
If this suggestion solved your problem, mark my post as a solution (blue box with a check mark) so that this ongoing discussion does not distract others who want to find out the answer to the original question and press the heart button. Thank you