Unable to deploy llm models in foundry

Question

Unable to deploy llm models in foundry

Alberto Jaimes 20

ASfter several attempts in a short time span; and can't raise a request in azure portal, can't advance in the menu

Update – exact error found

Azure portal > Monitor > Activity log shows every failed deployment (first attempts on Oct 25 and a retry today) with:

715-123420: "Our system has detected this request as unusual activity for your account. If you are confident this is in error, please contact support."

Context:

New subscription + Foundry resource, created for a Microsoft Learn exercise
Models tried: gpt-6-luna (default settings), grok-4.7, among others
Same error on every attempt

Other Q&A threads link this code to the automated fraud-protection (RTFP) hold, so I've stopped retrying and opened a free Subscription management request (ID: 2610270040000763); I can't open a Technical request without a paid support plan.

Could a Microsoft moderator please escalate this to have the hold reviewed? I can share the subscription ID and correlation IDs privately.

Thanks!

Nathan Roberts (SN) 14,196 Reputation points Student Ambassador Volunteer Moderator

2026-10-25T19:39:01.1566667+00:00

Hey there, Alberto Jaimes

When you try and deploy LLM models in Foundry, what error message do you get?

This will help the community try and help you with the problem you are facing.

Thanks,
Nathan

Answer accepted by question author

0 additional answers

Your answer

Nathan Roberts (SN) 14,196 Reputation points Student Ambassador Volunteer Moderator

2026-10-25T19:39:01.1566667+00:00

Hey there, Alberto Jaimes

When you try and deploy LLM models in Foundry, what error message do you get?

This will help the community try and help you with the problem you are facing.

Thanks,
Nathan

Answer 1

Alex Burlachenko 26,695 MVP Volunteer Moderator

Hi Alberto Jaimes & thx for join me at Q&A platform,

It's difficult to tell what's blocking the deployment from the information provided. The fact that you've made several attempts in a short period doesn't necessarily mean Azure has temporarily restricted your account.

I'd first check the exact error returned by the model deployment operation. In Microsoft Foundry, deployment failures can be related to regional model availability, insufficient quota, subscription restrictions, permissions, or service-side provisioning issues. Those require different fixes.

Please check Azure Portal > Monitor > Activity Log for the affected subscription and resource around the time of the failed deployment. If the operation appears there, the error details and correlation ID may help narrow it down. Also check the deployment error shown directly in Foundry, as not every failure is recorded in the subscription Activity Log.

Regarding the support request, when you say you can't advance in the Azure portal menu, is the Next button disabled, does the page fail to load, or are you being redirected to a different support option?

Make sure you're in the correct Microsoft Entra tenant and subscription. You can also try opening Help + support directly from the Azure portal in a private browser session to rule out a portal session issue. If the support workflow is blocked by permissions or subscription status, that needs to be addressed separately from the model deployment failure.

Could you share the model name, Azure region, deployment type (Global Standard, Data Zone Standard, etc.), and the exact deployment error or screenshot? Also, let us know at which step the support request stops.

Without those details, I wouldn't recommend repeatedly retrying the deployment or creating additional Foundry resources, as we don't yet know whether this is a quota, account, or backend issue

rgds,

Alex

p.s. if my answer is help - pls mark it as an answer

Alberto Jaimes 20 Reputation points

2026-10-27T01:46:15.48+00:00

Hello Alex, good afternoon, thanks for answering

Long version, I created a new subscription group with a Foundry resource to complete an exercise from an microsoft learning module, when I tried to deploy an LLM model with default settings (gpt-6-luna), it failed, then tried again with the same model, then tried again with other models (e.g. grok 4.7), ..

Going to Azure portal > Monitor > Activity log allowed me to recover the error I saw at the time, error code 715-123420, shown in the attache image:

Today I tried again to deploy a model in Foundry, getting the message: Model deployment failed; but checking it in Azure portal > Monitor > Activity log I see the same message.

I couldn't advance in raising the support request in the azure portal because the suggested options were irrelevant, then it suggested me to buy a support package. Now that I have an error code, I will try to raise a support request again.

Problem is not solved, but I mark your message as an answer.
Alberto Jaimes 20 Reputation points

2026-10-27T01:57:08.2933333+00:00

Hello Alex, good afternoon, thanks for answering

Long version, I created a new subscription group with a Foundry resource to complete an exercise from an microsoft learning module, when I tried to deploy an LLM model with default settings (gpt-6-luna), it failed, then tried again with the same model, then tried again with other models (e.g. grok 4.7), ..

Going to Azure portal > Monitor > Activity log allowed me to recover the error I saw at the time, error code 715-123420, shown in the attache image:

Today I tried again to deploy a model in Foundry, getting the message: Model deployment failed; but checking it in Azure portal > Monitor > Activity log I see the same message.

I couldn't advance in raising the support request in the azure portal because the suggested options were irrelevant, then it suggested me to buy a support package. Now that I have an error code, I will try to raise a support request again.

Problem is not solved, but I mark your message as an answer.