Model not found" error when deploying custom model in Azure AI Foundry Body:

Question

Model not found" error when deploying custom model in Azure AI Foundry Body:

zhuzin zhuzin 20

Hi everyone,

I recently trained a custom text classification model using Azure AI Foundry and tried deploying it as an endpoint. The training completed successfully, but when I attempt to test or call the endpoint, I get the following error:

ModelNotFound: The specified model could not be found or is not accessible in this region.

Answer accepted by question author

0 additional answers

Your answer

Answer 1

Hi there

usually happens when the model registry and the deployment endpoint are not properly linked within the same Azure AI hub. Even though both appear in the same region, sometimes deleting and recreating the hub or resource group can break the internal linkage.

Try these steps:

Go to your Azure AI Foundry → Project → Assets → Models, click on the model, and verify that it’s registered under the same hub and subscription as your endpoint.

If it looks fine, try re-registering the model manually using the Foundry CLI or SDK, then redeploy.

Also make sure the model type (custom, foundation, or fine-tuned) matches the deployment configuration — Foundry can reject mismatched types with a “Model not found” error.

Finally, check if your region (East US) has the quota or feature enabled for custom model deployments; some model types are region-limited.

If none of these help, try deploying the model through the Foundry Playground instead of API — if that works, the issue is likely configuration-related in your deployment script.

Share via

Model not found" error when deploying custom model in Azure AI Foundry Body:

0 additional answers

Your answer