Unable to Assign GPT-4o Models After Upgrade – Support Request

Question

Hello Microsoft Support Team, I am writing to report an issue with my Azure OpenAI model access following a recent account upgrade. Despite having upgraded my subscription/tier, I am still unable to assign or deploy GPT-4o models in Azure AI Foundry. The models do not appear to be available for deployment in my environment. Details: - Issue: GPT-4o models unavailable for assignment/deployment post-upgrade - Platform: Azure AI Foundry / Azure OpenAI Service - Subscription ID: [ 73b41168-254c-456c-9aaa-1ef99ffd0b15  ] - Region: [US EAST2] - Upgrade completed on: [28/03/2026] Steps I have already taken: - Confirmed the upgrade was successful - Checked model availability in my region - Reviewed quota limits in the Azure portal Could you please investigate why GPT-4o model deployment access has not been provisioned following my upgrade, and advise on next steps? Thank you for your assistance. Best regards, [Ali Emadi]

Answer 1

Hello Ali Emadi,
Upgrading your Azure subscription (Pay-As-You-Go, Sponsorship, EA→MCA, Free→Paid, or S0 tier) does not automatically grant access to GPT-4o deployments in Azure AI Foundry. GPT-4o availability is controlled by a separate model allow-listing entitlement, not billing status. So even if billing is upgraded, the region supports GPT-4o (e.g., East US 2), and quota is visible, the model may still not appear because your subscription hasn’t yet been added to the GPT-4o access policy.

In this case, the subscription is registered for Azure OpenAI, billing is enabled, the resource provider is active, and quota is allocated — but GPT-4o allow-listing is still missing. Azure AI Foundry only shows models your subscription is authorized to deploy, not all models supported in a region. This often happens after upgrades because entitlement sync runs asynchronously and may delay or miss subscriptions, causing GPT-4o to be silently hidden.

To confirm, run az cognitiveservices account list-models in Azure Cloud Shell. If gpt-4o or gpt-4o-mini don’t appear, it confirms the issue is missing entitlement, not quota.

Answer 2

Hi Ali,

Thanks for reaching out here.

The GPT-4o models are not appearing in your Azure AI Foundry environment because they are in the process of being retired and replaced by GPT-5.1. Even though you upgraded your subscription, the retirement schedule means GPT-4o deployments are being phased out starting March 2026, with auto-upgrades already underway. Microsoft has designated GPT-5.1 as the successor to GPT-4o. Subscription upgrades do not override retirement schedules. Even with higher tiers, deprecated models are removed from catalogs.

Please consider the following recommended next steps:

1. Check model catalog - Navigate to the Azure AI Foundry Model Catalog to confirm which models are currently available as replacement. You can refer here - https://jinlee794.github.io/foundry-model-availability-notifications/models/gpt-4o/
2. Deploy GPT-5.1 instead - Begin migration to GPT-5.1 for continuity. Microsoft provides auto-upgrade paths, but manual redeployment ensures you retain control over configuration. Please note that GPT-4o will no longer be assignable after retirement dates. Attempting deployment will fail, regardless of subscription tier.
3. Verify Quotas - Confirm that your subscription tier includes quota for GPT-5.1 deployments.
4. Monitor Retirement Notices - Keep track of official Azure OpenAI retirement documentation for updates on timelines and replacement guides.

If GPT-5.1 is not visible in your catalog despite retirement notices, escalate via Azure Support with your subscription ID and region details.

Please note that if you had existing GPT-4o deployments, they may already be auto-migrated to GPT-5.1 - you need to verify your endpoints to avoid unexpected changes. Some regions may show delayed catalog updates. If GPT-5.1 is not yet listed, Microsoft support can confirm rollout status.

You can read more on this here:

https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirements?tabs=text

Hope this was helpful! Let me know how it goes and if you have any further queries in this regard.