Azure OpenAI in Microsoft Foundry Models Quotas and Limits in Azure Government

This article features detailed descriptions and best practices on the quotas and limits for Azure OpenAI in Azure Government.

This article provides a quick reference and a detailed description of the quotas and limits for Azure OpenAI in Azure Government.

Scope of quota

Azure Government doesn’t enforce quotas and limits at the tenant level. Instead, it enforces the highest level of quota restrictions at the Azure subscription level.

Regional quota allocation

Tokens per minute (TPM) and requests per minute (RPM) limits are defined per region, per subscription, and per model or deployment type.

For example, if the gpt-4.1 DataZone Standard model has a quota of 5 million TPM and 5,000 RPM, then each region where that model or deployment type is available has its own dedicated quota pool of that amount for each of your Azure subscriptions. Within a single Azure subscription, you can use a larger quantity of total TPM and RPM quota for a given model and deployment type, as long as you spread resources and model deployments across multiple regions.

Quota tiers

In Azure Government, the service doesn’t support quota tiers or automatic adjustments to quota. Instead, it provides two levels: a default level and an enterprise level for customers with an Enterprise Agreement.

Can I request more quota?

Yes. Use the Azure Government quota request form to request both standard quota and fungible provisioned throughput unit (PTU) quota. If a standard quota request is approved, the current tier stays the same, but you get more quota.

Azure Government quota reference

DataZone Standard

DataZone Standard Tokens Per Minute (TPM)

Model name	Default TPM	Enterprise TPM
gpt-5.6-sol	1,000,000	2,500,000
gpt-5.6-terra	1,000,000	2,500,000
gpt-5.6-luna	1,000,000	2,500,000
gpt-5.1	300,000	1,000,000
gpt4.1	300,000	2,000,000
gpt-4.1-mini	300,000	2,000,000
gpt-4o	300,000	10,000,000
o3-mini	2,000,000	2,000,000

Standard

Standard Tokens Per Minute (TPM)

Model Name	Default TPM	Enterprise TPM
gpt4.1	300,000	2,000,000
gpt-4.1-mini	300,000	2,000,000
gpt-4o	80,000	80,000
text-embedding-3-large	350,000	350,000
text-embedding-3-small	350,000	350,000
text-embedding-ada-002	350,000	350,000

General best practices to stay within rate limits

To minimize problems related to rate limits, use the following techniques:

• Implement retry logic in your application.
• Avoid sharp changes in the workload. Increase the workload gradually.
• Test different load increase patterns.
• Increase the quota assigned to your deployment. Move quota from another deployment, if necessary.

Regional quota capacity limits

You can view quota availability by region for your subscription in the Foundry portal.

To view quota capacity by region for a specific model or version, query the capacity API for your subscription. Provide a subscriptionId, model_name, and model_version. The API returns the available capacity for that model across all regions and deployment types for your subscription.

Currently, both the Foundry portal and the capacity API return quota and capacity information for models that are retired and no longer available.

Related content

• Explore how to manage quota for your Azure OpenAI deployments.
• Learn more about the underlying models that power Azure OpenAI.
• Learn about provisioned throughput in Azure Government.