Azure ML Deployment Fails with Out‑of‑Region Quota Error Despite Available A100 Quota (Central US)

Hi there,

While creating a new deployment in our Azure Machine Learning workspace (Central US region), we encountered this error: ResourceOperationFailure: Out of region quota for the requested VM size. Please refer to the troubleshooting guide here: https://aka.ms/oe-tsg#error-outofquota.

We have a quota of 192 Standard_NC24ads_A100_v4, which allows us to create 8 GPU instances. Since Azure Machine Learning reserves 20%, we’re able to create up to 6 GPU instances. Currently, we have 3 existing A100 instances, and when we tried to create 3 more with this deployment, we received the error above. Do you have any suggestions on how to resolve this?