Finetuning Server issues


While finetuning I keep getting the following error:
Server error. Returning to queue for retry
Exceeded allowed number of retries. Fine-tune failed. For help, please contact OpenAI and include your fine-tune ID:

Could this be an issue with the finetune training file I uploaded?



I’m getting the same issue today with finetunes. All are failing to launch.

1 Like

I have been getting the same error message for my last couple of finetune attemps. I’m also wondering if there’s something wrong with the training file (.jsonl format) or if there’s an issue on the server side? My training data only contains language detection prompts. So maybe there’s an issue with Chinese or Japanese characters?

Does anyone have any new information on this topic?
This is the output from the fine-tuning stream:

Server Error on Finetuning

My fine tune succeded at last, so I guess it wasn’t an issue with the training file.


I have the same problem.
How did you solve it ?



I have the same problem too. I first thought that my dataset or its format is somewhat collapsed something like ‘{’, but the file is correct. It is not because of Japanese or Chinese characters since I had used it in my first fine-tuning try. Now I have trying to run fine-tuning task again, and it seems to work.

(Still, I can’t find the reason why the last task failed.)

I found that my account cause a rate limit error.
Not only Fine-tuning but also normal using of models that already tuned doesn’t work, and we recieve:

"openai.error.RateLimitError: Your account is not active, … "

Error says that I should check the billing status but it might not be the matter, just
it seems that the use of all of our organization reaches the too frequential use.

However, who uses in high frequency in my organization? No-one here but I am only the user of our teams.
Finally I can’t find the solution.