Don't have quota to do a model deployment in Foundry

lakshmi vishnu vardhan reddy sontem 5 Reputation points
2026-06-08T15:28:08.48+00:00

Hello,

I recently created both azure and foundry accounts. I am trying to deploy a gpt-4.1 and gpt-4.1mini model under a new project to do microsoft module practice exercises but when i try to deploy, it says insufficient quota to all regions. So I checked at quota section under project details and there are 0 quotas available. How do I request quota to my project?

Thanks

Community Center | Not monitored
0 comments

1 answer

Sort by: Newest
Most helpful Newest Oldest

  1. AI answer

    2026-06-08T15:28:26.76+00:00

    To get quota so model deployments work in a new Foundry project, use both the Foundry portal quota tools and the quota request form.

    1. Verify roles on the subscription
      • Ensure the account has the required roles on the Azure subscription:
        • Cognitive Services Usages Reader to view quota.
        • Owner or Contributor to request quota increases.
        • Cognitive Services Contributor + Cognitive Services Usages Reader to edit allocations in Foundry.
      • Without these, the Quota page may show no allocations or the Request quota button may be disabled.
    2. Check and manage quota in Foundry (new)
      • Go to Microsoft Foundry and ensure New Foundry is enabled.
      • Select the correct project (upper-left corner).
      • Select Operate (top navigation).
      • Select Quota in the left pane.
      • Use the Token per minute tab to view TPM quota for standard deployments.
      • If any deployments exist that are using quota, select them, open the details pane, and in Affiliated deployments using shared quota use the pencil icon to reduce or reallocate TPM so that quota is freed for the new gpt‑4.1 / gpt‑4.1‑mini deployments.
    3. Request more quota from within Foundry
      • On the Quota page for the project, select Request quota (upper-right corner).
      • Submit a request for the models and regions where gpt‑4.1 and gpt‑4.1‑mini are needed.
      • Allow up to ~15 minutes after approval for changes to propagate, then refresh the Quota page and retry deployment.
    4. Request quota via the central form (if no quota at all)
      • If the subscription has 0 quota for these models in all regions (not just unallocated), submit the global quota request form: quota increase request form.
      • Use this to request quota for Azure OpenAI models (including gpt‑4.1 and gpt‑4.1‑mini) in the desired regions.
      • Quota requests are processed in order and are more likely to be approved if existing quota (if any) is actively used.
    5. If quota UI shows issues
      • If the Quota page is empty or shows no allocations even after roles are confirmed, verify the correct subscription is selected.
      • If the Request quota button is disabled, confirm Owner or Contributor role and that the model/region supports quota increases.
      • If quota changes are approved but not visible after 24 hours, open an Azure support ticket as advised in the troubleshooting guidance.

    Once quota is granted and visible on the Quota page, retry deploying gpt‑4.1 and gpt‑4.1‑mini in a region where TPM is available.

    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.