The error “We’re having trouble accessing your files right now. Please try again later.” occurs at the fine-tuning endpoint

The community suggests using the gpt-4.1-mini model, but is there no way to successfully tune gpt-4o-mini?

I encountered the same issue. Is it still possible to use 4o-mini normally? Has the official team disabled the use of this model?

The interesting thing is that if I try about 100 times, I succeed once.

Hi crew - here from the OpenAI team. We took a look into these reports and found out there was an issue on our side that has now been fixed. Can you try again and let me know how it goes?

5 Likes

Thank you for both the fix and the communication! I can confirm it started succeeding for me at 1pm EDT 3/18/2026.

1 Like

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.