no…
It still remains occurring irregularly.
March 6th, 2025 - I am experiencing this bug today as well. 10 batch jobs, all stuck in validating for 4+ hours. Cancelling looks to be stuck as well. Restarted job that kicks off batch and 4 went right through, 6 got stuck in validating again.
Likewise, this issue seems to have reappeared at roughly Thu, 06 Mar 2025 12:12:41 GMT (1741263161 in epoch time).
same here, stuck in validating for long with o3-mini batch request, canceled and retried for times and not working…
I encountered the same issue today, with all of my batch jobs failing to process. Is this problem limited to specific models? If anyone has any tips or insights to make a batch go through, they would be greatly appreciated.
I have opened a thread with support and will update here as soon as I have more information from them.
I restarted my job a third time and 0/10 have moved beyond the validation phase, so something is definitely up. My jobs are on gpt-4o
1 Like
There was an internal issue causing batches to not get validated for a while. We have fixed it and you should see your batches pass the validation phase very soon. Apologies for the trouble!
3 Likes
Thanks! Just came here to comment that it looks fixed for me too.
Hello! Throughout the day I was running some queries via the Batch API (on /v1/responses endpoint) and didn’t have any issues until 3:30-4pm today. Since then all my requests are stuck at the validating state, and cancelling/rerunning does not help. https://status.openai.com/ does not show any outage for the Batch API. I didn’t want to open a new thread for the same problem. Does anybody experience similar delays today?