Assistance Needed for Azure OpenAI Quota Increase

Question

Assistance Needed for Azure OpenAI Quota Increase

pipi 30

We would like to request an Azure OpenAI quota increase to match the current limits provided to us by OpenAI.

Our current TPM limits with OpenAI are:

gpt-4.1-mini: 150,000,000 TPM

gpt-4.1: 30,000,000 TPM

We already have a production environment running on OpenAI and would like to use Azure OpenAI as a backup in case OpenAI’s service experiences degradation. To ensure service continuity, we need the same quota in Azure as we currently have with OpenAI.

We submitted the quota increase form last week, but Microsoft only approved 5,000,000 TPM for gpt-4.1-mini. When we requested a further increase, we were advised to either contact a Microsoft Representative or create a support ticket.

Could you please advise:

What is the correct process to request a higher quota (150M TPM) for gpt-4.1-mini?

Which regions and subscription tiers would allow this higher quota?

0 comments

1 answer

Your answer

Answer 1

Pavankumar Purilla 11,655 Moderator

Hi pipi,
To request a significantly higher quota such as 150M TPM for gpt-4.1-mini, the correct process is to submit a support request through the Azure Portal by selecting Service and subscription limits (quotas) → Azure OpenAI Service, and then choosing the specific deployment region. In the details of your request, please include your current usage limits with OpenAI, the intended use case for Azure as a backup environment, and any supporting information on expected sustained throughput requirements. Requests of this scale require review by our Capacity and Quota Management team and may also require coordination with your Microsoft account team. If you do not already have a dedicated Microsoft representative, our support engineers can help facilitate this escalation.

Please note that quotas of this size are typically approved only for customers on Enterprise-level agreements (Enterprise Agreement or Microsoft Customer Agreement – Enterprise tier) and may only be available in regions with sufficient capacity, such as East US, South Central US, West Europe, or Sweden Central. Capacity availability can vary by region, so part of the review process will involve confirming that the requested throughput can be supported in your chosen deployment location.

For reference:
Azure OpenAI in Azure AI Foundry Models quotas and limits
Manage Azure OpenAI in Azure AI Foundry Models quota
Create an Azure support request

pipi 30 Reputation points

2025-08-13T03:55:24.77+00:00

Hi Pavankumar,

Thank you for the reply. I have already submitted the request quota via Azure Portal last week, but Microsoft only approved 5,000,000 TPM for gpt-4.1-mini on Japan East.

So it is more recommended that i request for East US?
Pavankumar Purilla 11,655 Reputation points Moderator

2025-08-14T05:09:01.02+00:00

Hi pipi,
Since Azure OpenAI quota availability varies by region, very high TPM allocations such as 150M are generally approved only in regions with the largest capacity, often referred to as “Data Zone Enterprise” regions. Japan East has more limited capacity, which is why the approved quota was lower. For higher limits, we recommend submitting your quota request for a region with greater availability, such as East US, South Central US, or West Europe. These regions typically have more capacity to accommodate large-scale allocations, especially for enterprise customers. You can also consider deploying in multiple high-capacity regions to improve redundancy and meet your failover requirements.
Anonymous

2025-08-20T04:29:42.0666667+00:00

Hi pipi,

Did you get any chance to check the response.

Thank you!