getting "715-123420: An error occurred. Please reach out to support for additional assistance" error while deploying Claude Sonnet 4.6 model

Question

getting "715-123420: An error occurred. Please reach out to support for additional assistance" error while deploying Claude Sonnet 4.6 model

Answer 1

Hi User1-60039231

Thanks for sharing the details.

This error code 715‑123420 is something we have seen in a few recent Azure AI Foundry and Azure OpenAI deployment scenarios. In most cases, this error is not caused by an issue in your template, code, or configuration.

What usually happens is that the deployment request gets blocked by a backend safety or fraud protection check. This can happen even when the subscription, quota, and permissions all look correct. Because of that, retrying the same deployment in the same setup often continues to fail.

A few things you can try to narrow this down:

First, try deploying the same model in a different supported region. Many users have reported that the deployment works in regions like East US while failing in others, which points to a regional backend condition rather than a subscription issue.

Second, if this is not a fine‑tuned model, try creating a new deployment under a new or existing Foundry project and deploy the model again. In several cases, creating a fresh deployment cleared the error.

Third, double‑check that there are no network restrictions such as private endpoints or custom networking blocking the request. For testing, using the default public network access can help confirm whether networking is part of the problem.

This error is not currently documented as a standard ARM or Foundry configuration error, and it does not usually indicate quota exhaustion or missing permissions. It is typically a temporary or backend‑enforced block that affects certain deployments or regions.

For reference, Microsoft explains how model deployments and alerts behave in Azure AI Foundry here https://learn.microsoft.com/azure/ai-services/openai/overview

And model and region availability can be checked here https://learn.microsoft.com/azure/ai-services/openai/concepts/models

If you retry in a different region or create a fresh deployment and the issue clears, that confirms it was not related to your setup. Please let us know what you observe after trying the above so the community can help further.

If this answers your query, please do click Accept Answer and Yes for was this answer helpful.

Thankyou!