We are encountering error code 500 repeatedly when using the OpenAI API. Our servers in both the US and Tokyo are affected. These prompts had been working without issue for weeks, but over the past two days the batch job began failing part-way through with 500 errors for all tasks. Even previously (5 mins before this problem is encountered) successful prompts, when re-executed, immediately return error 500. We have not changed the prompt or request structure.
1 Like
Same here, I also reported the 2 days window. Critical bug using Flex model for GPT5 since last Monday - 500 Internal server error
2 Likes
I also get huge number of 500 errors, when near identical batches used to process fine. And i see problems with openAI’s count of the errors
Same here. All my requests that used to work were all failing with 500s using GPT-5 in the flex tier.