unable to deploy GPT-5.2 modules in Azure Foundry like training says to do

Andrew Laporte 0 Reputation points
2026-07-27T15:22:50.27+00:00

Hi,

I'm trying to follow the MS Azure Foundry training for AI-103. In the first exercise it says to create a Azure subscription, create a foundry project and then add a model to the project.

https://microsoftlearning.github.io/mslearn-ai-studio/Instructions/Exercises/01-Explore-ai-studio.html

I keep getting an error when trying to add gpt-5.2 to the project that I'm out of quota or resources but it does not say what to do if that happens.

The error is "gpt-5.2 isn't available due to insufficient quota. Choose a different model or deployment option for gpt-5.2. You can also request more quota.

I have no idea how to request more quota. I can use some help in getting started here.

Andy

Azure OpenAI in Foundry Models
0 comments No comments

1 answer

Sort by: Most helpful
  1. AI answer

    2026-07-27T15:23:12.7833333+00:00

    gpt-5.2 is showing insufficient quota because the subscription does not currently have Token-per-Minute (TPM) quota allocated for that model in the selected region.

    To get started:

    1. In Azure AI Foundry, open the quota area for the subscription and region planned for deployment.
    2. If gpt-5.2 does not appear in quota management, submit the manual quota request form: https://aka.ms/oai/stuquotarequest.
    3. In the request, specify:
      • the Azure subscription
      • the target region
      • the model name: gpt-5.2
      • the requested TPM amount for a small test deployment
    4. After the request is approved, the quota should appear in Foundry and the deployment can be created.

    If quota already exists for the model, another supported option is to reduce or reallocate TPM from other deployments in the Foundry portal to free capacity.

    If deployment still fails after quota is granted, verify that:

    • gpt-5.2 is available in the selected region
    • the deployment has enough quota assigned
    • the deployment reaches provisioningState: Succeeded

    A quick alternative for the exercise is to deploy a different available model if the training does not strictly require gpt-5.2.


    References:

    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.