Gemini Ai Fine Tuning problem

I’ve been experiencing difficulties with Fine-Tuning any model in the Google AI Studio. It seems to be stuck &, showing “in queue” and “in progress” without any change. This issue has persisted for the past 3 to 5 days. Are others encountering the same problem? When can we expect it to be resolved?

1 Like

Yes, others have experienced the same issues:


Not sure, Google is currently aware of the issue.

I had the same problem. Apparently you can only fine tune 500 examples. But thats not the issue at all. Its much easier to do it on Vertex AI under language/fine tuning.

Hello! is the problem still happening today? a fix went in yesterday

No asoroken ,Now its working Thank you :slight_smile:

1 Like

yeah ,but the cloud is not accepting any of my cards .so I am unable to get the free credits

The corpus used to fine-tune the model contains units such as 10¹² and m³, which become 1012 and m3 after being input into the corpus. How should this be handled?

Ok even with Vertex AI, i am getting the follwong error message. I have tried to contact google support. They acknowledge the problem. but still no return.

com.google.cloud.ai.platform.common.errors.AiPlatformException: code=RESOURCE_EXHAUSTED, message=The following quota metrics exceed quota limits: aiplatform.googleapis.com/custom_model_training_cpus, cause=null; Failed to create custom job for the task.

It appears that your error specifically is telling you that you have surpassed your usage quota during the API call. There’s not much that can be done there except to wait for your quota to refresh.

Hello, I still cannot upload a list of 20 examples in google sheets or csv format. Once I upload the file it’s like nothing was uploaded and I cannot click on “Tune”