This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
Hi everyone,
I am working on a project where we have to create a Foundry account, Foundry project and deploy Anthropic model Claude Opus 4.6 (GlobalStandard, 40K TPM).
I have the Contributor role in a new PAYG (Pay-as-you-go) Azure subscription and Marketplace access is enabled for the subscription. I navigate to Microsoft Foundry web site (https://ai.azure.com/<tenant>) and create a project in Azure region Sweden Central. Then, I clicke on Discover, find the Anthropic model Claude Opus 4.6 and I start the model deployment using the default settings. After a few minutes the model deployment reports status Failed without giving me any more context with regards to the exact reason it failed.
I have tried the exact same thing on another subscription and it works without any issues.
I am lost. Can anyone please help me locate the issue?
A catalog of AI models in Microsoft Foundry that you can discover, compare, and deploy using Azure’s built‑in tools for evaluation, fine‑tuning, and inference
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(176, 16%, 27%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESC%3C/text%3E%3C/svg%3E)
We are currently investigating the issue with the internal team. Please allow us some additional time, and we will share an update with you as soon as we have more information.
Thank you!
Hi, did you find anything with regards to my issue? Is there anything I can do to help?
Thank you for the additional details.
What Yutaka mentioned is correct for Anthropic Global Standard deployments, there are two separate layers involved:
The backend subscription/regional quota allocation required by the Azure OpenAI service
In some cases, the Foundry UI may display available TPM quota, but the backend provider enablement or regional capacity allocation for the subscription may not yet be fully provisioned. This can result in deployments failing without a detailed validation message.
Since:
The same deployment succeeds in another subscription
Marketplace access is already enabled
You had visible quota prior to deployment
The quota changes after deployment attempts
this may indicate a backend quota allocation or provider onboarding issue for the subscription.
At this stage, we are checking this internally with the engineering team to verify:
As a temporary workaround, you may also test:
A smaller TPM deployment
Another supported region
Recreating the deployment after a few minutes in case quota propagation is still in progress
We’ll investigating this from our side and share updates as soon as we have more information.
Thank you!
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(156.8, 57.00000000000001%, 25%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHD%3C/text%3E%3C/svg%3E)
Hello,
I'm having the same issue across multiple subscriptions. I think the details are very similar to @Panagiotis Tzedakis :
For what it's worth, this appears to be happening with multiple Claude models.
I'm definitely interested in what you're able to find out - I'm stuck without any detailed error logging.
Thank you!
Haig Didizian
Clearistic Technology
Did you get any chance to review the above response. Do let me know if you have any further queries.
Thank you!
I think region, Contributor on the Foundry resource group, the Anthropic Marketplace offer being allowed, and that the subscription can allocate the Anthropic GlobalStandard quota. If the quota is zero, the deployment just fails with no extra detail.
Hi Yuma and thanks for your comment.
At the time of creating the Microsoft Foundry resource there was available the whole 40K TPM quota to consume, so I do not think this is why the Microsoft Foundry deployment failed.
I see, thanks. Just to clarify — 40K TPM shown in Foundry is the project‑side limit. The Global quota is configured in the Azure subscription per region, and deployments can fail if the required subscription quota is not available. Maybe worth checking the subscription quota.
I see. I did not know that. How can I check the subscription quota?
I checked the quota and the thing is that before I deployed the model in the specific region, I had all the quota available (40K TPM). When I attempted to deploy the model, the deployment failed but the subscription quota for the model now shows 0 TPM available, which I feel makes sense. However, I still do not understand why the deployment gets failed. It does not even log a message to help me understand the failure reason. It is strange.
Global quota only inside Foundry → Management Center → Quota. If it shows 0/0, the subscription has no assigned capacity, and Azure support handles the “Anthropic Global Standard enablement” request.
Thank you. @SRILAKSHMI C Can you check what Yutaka says in case this is something you need to handle on your end? My client needs to have the model setup asap and I do not know what else to do.
Thanks. Yes — this is a subscription‑side GlobalStandard enablement. SRILAKSHMI can assign the quota on their side.