I am constantly getting this error - RuntimeWarning: coroutine 'FineTune.stream_events' was never awaited cls._stream_events(resp["id"])

I am getting this error for a few days, earlier I was able to fine-tune the model successfully.

3 Likes

From what I see, it’s saying “client disconnected…”

Did this Notebook work before?

Did you try to resume?

After resuming also it is giving the same error.

Latest version of the OpenAI package?

I am getting the same error, start from today, before the fine-tuning with the same configuration worked nice. Is there something was updated?

Might be. I’m not sure from the outside.

Status page shows some problems last few days, though…

Might be best to reach out to support or give it a while and try again.

yep, and it is also debiting money inspite of giving the error which is worst I guess. I lost around $70 while fine-tuning because of this error.

1 Like

Ouch. I’d definitely reach out to support if you haven’t yet, and let them know. Hopefully they refund for you.

I have the same issue. While it gives this error, it seems the model creation is working. It creates the model, and when you run “openai api fine_tunes.list” everything appears to be in order. It seems something is wrong with the streaming of the results.

1 Like

It seems to be working with OpenAI 0.25 and not with the current version 0.26
Try

pip uninstall openai
pip install openai==0.25.0
6 Likes

This worked for me. Maybe some issue with the latest update?

1 Like

I’m getting the same error. I updated to the latest version a few hours ago before starting the finetuning process.

1 Like

I am facing the same issue. It gives an error while submitting a fine-tuned request, but when I check fine-tuned model list, it has created six fine-tuned models wasting my money.

Same here, did you ask them for a refund?

This is getting solved, thanks to the dev team, see here:[0.26.0] openai/cli.py:440: RuntimeWarning: coroutine 'FineTune.stream_events' was never awaited · Issue #173 · openai/openai-python · GitHub

1 Like

Bless you. Hope you stick around the forums. We could use more people like you! :wink:

I didn’t even think to check GitHub comments, but that’s likely a better place than here for the devs to see the probs…

Thanks again.

1 Like

Guys, please help me in this post too, I am not able to handle the context situation in GPT3; so, if you have some solution to this then it will be very helpful for me.