Azure Foundry - Error deploying model

Question

I have been trying to deploy GPT-5.4-mini in Azure Foundry but I keep getting the following error:

Failed to create deployment An error occurred. Please reach out to support for additional assistance.

Answer 1

Hi Sebastian,

Sorry you're running into this issue.

The error message "Failed to create deployment. An error occurred. Please reach out to support for additional assistance." is fairly generic and doesn't identify the root cause on its own.

A few common things to check are:

1. Verify model availability

Ensure the model you're trying to deploy is available in your selected region and is supported for your subscription.

2. Check quota availability

Insufficient quota is a common reason for deployment failures. In Azure AI Foundry, navigate to Quota and verify that you have sufficient capacity available for the model and region you're deploying to.

3. Review deployment settings

Confirm that the selected model version, deployment type, and other configuration settings are valid for your scenario.

4. Check deployment details and logs

If available, review the deployment operation details in Azure AI Foundry or Azure Portal for a more specific error message. The generic banner often hides the underlying cause.

5. Retry the deployment

Occasionally, transient service issues can cause deployment failures. Refresh the portal and try the deployment again.

To help narrow this down, could you provide:

• The model you're trying to deploy
• The Azure region
• Whether you're deploying through Azure AI Foundry, SDK, CLI, or REST API
• Any additional error details shown in the deployment operation logs

With those details, the community may be able to identify the exact cause more quickly.

Reference:

Create model deployments in Azure AI Foundry

How to deploy and infer with a managed compute deployment (classic)

I Hope this helps. Do let me know if you have any further queries.

Thankyou!

Answer 2

Hello Sebastian Quiroga,

Greetings! Thanks for raising this question in Q&A forum.

The exact wording you are seeing, "An error occurred. Please reach out to support for additional assistance," matches a specific and already known issue rather than a typical quota or region misconfiguration. This message is tied to error code 715-123420 (sometimes shown as 751-123420), and several other people deploying GPT-5.4 family models, including gpt-5.4-mini specifically, have hit this exact same code recently. According to Microsoft moderators tracking these reports, this code means the deployment request is being blocked at the platform level by Azure's automated risk and fraud protection screening (RTFP), not by your quota, region, or configuration. That is why it can happen even when quota is visible and other models deploy fine on the same subscription.

Here is what I would do.

Confirm the error code Open the deployment attempt again and check if a code like 715-123420 or 751-123420 appears alongside the message, either in the portal error banner or in the deployment operation details. This confirms you are hitting the same known issue rather than something else.

Do not keep retrying repeatedly from the portal Since this is a backend platform level block rather than a transient service hiccup, retrying the same deployment over and over will not get past it.

Open a support ticket specifically for this code Go to Help + Support in the Azure portal and open a technical support request under your Foundry resource's subscription, referencing error 715-123420 and the GPT-5.4-mini deployment attempt. Include your subscription ID, the Foundry resource and project name, the region you are deploying to, and the timestamp of a recent failed attempt. This code requires Microsoft to create an internal collaboration ticket with the security team to manually review and unblock the subscription, it is not something that resolves itself from retrying or changing settings.

While waiting, sanity check the basics It is still worth quickly confirming the model is listed as available in your chosen region and that your subscription type supports it, just to rule out a simpler cause while your ticket is being reviewed, but based on the pattern matching your error message, the fraud protection block is the most likely explanation.

If this answer helps you kindly accept the answer which will help others who have similar questions

Best Regards,

Jerald Felix.