Microsoft Foundry Models quotas and limits

This article provides a quick reference and detailed description of the quotas and limits for Foundry Models sold by Azure. For quotas and limits specific to the Azure OpenAI in Foundry Models, see Quotas and limits in Azure OpenAI.

Subscription-level quota management

Starting with Realtime Translate and Realtime Transcribe, and soon all models, Foundry tracks quota for deployments at the subscription level rather than per resource or per region. This approach brings consistency and predictability to how quota is managed across deployments, since all resources and regions in a subscription share the same quota pool.

This change consolidates quota into shared pools:

• Global Standard: Deployments of the same model and version share one quota pool across all regions in a subscription.
• Data Zone Standard: Deployments of the same model and version share one quota pool per data zone (for example, US or EU).

Quota tiers

Quota tiers help Foundry Models workloads scale by increasing quota automatically as usage grows. Seven tiers are available: Free Tier and Tiers 1 through 6, with Tier 6 providing the highest quotas. Initial tier assignment considers usage and the customer's relationship with Microsoft, including enterprise agreements such as EA and MCA-E.

What's changing for me?

Evaluate quota-tier eligibility at the Azure subscription level. Eligibility can depend on usage, payment history, and the subscription's Azure offer. Some models and deployment types also have offer-specific quota limits.

Some models require a higher quota tier before you can deploy them. If your subscription doesn't meet a model's tier requirement, use a model available at your current tier while your subscription becomes eligible for an upgrade.

You can request additional quota without changing tiers. If the request is approved, assign the additional quota while your subscription remains in its current tier.

How does my subscription move to a higher tier?

Automatic upgrades primarily consider consumption trends across Foundry Models over time. As usage grows and the current tier constrains that usage, eligible subscriptions can advance to a higher tier.

Eligibility also considers the customer's relationship with Microsoft and payment history. Enterprise relationships, including EA and MCA-E agreements, are factors in assigning higher tiers.

Can I opt out of automatic upgrades?

Yes. You can keep your subscription in its current tier by setting its tier upgrade policy to NoAutoUpgrade.

Some customers use quota to constrain spending. However, quota isn't the recommended tool for managing costs. For guidance, see Cost management.

Before you run the Bash example:

• Install the Azure CLI and sign in with az login.
• Replace {YOUR-SUBSCRIPTION-ID} with your Azure subscription ID.

The following command sets the subscription's tier upgrade policy to NoAutoUpgrade:

subscription_id="{YOUR-SUBSCRIPTION-ID}"
az rest --method patch \
  --url "https://management.azure.com/subscriptions/${subscription_id}/providers/Microsoft.CognitiveServices/quotaTiers/default?api-version=2025-10-01-preview" \
  --headers "Content-Type=application/json" \
  --body '{"properties":{"tierUpgradePolicy":"NoAutoUpgrade"}}'

Reference: az rest, Quota Tiers - Update.

Can I request more quota?

Yes. Submit the quota increase request form. If your request is approved, your subscription receives additional quota but stays in its current tier. A quota increase isn't a tier upgrade.

How do I check my subscription's quota tier?

Use the Quota Tiers - Get REST API to retrieve the subscription's current tier and upgrade policy.

import json

import requests
from azure.identity import DefaultAzureCredential

subscription_id = "{YOUR-SUBSCRIPTION-ID}"
api_version = "2025-10-01-preview"
url = (
    f"https://management.azure.com/subscriptions/{subscription_id}"
    "/providers/Microsoft.CognitiveServices/quotaTiers/default"
    f"?api-version={api_version}"
)

with DefaultAzureCredential() as credential:
    token = credential.get_token("https://management.azure.com/.default")
    response = requests.get(
        url,
        headers={"Authorization": f"Bearer {token.token}"},
        timeout=30,
    )
    response.raise_for_status()
    print(json.dumps(response.json(), indent=2))

subscription_id="{YOUR-SUBSCRIPTION-ID}"
az rest --method get \
  --url "https://management.azure.com/subscriptions/${subscription_id}/providers/Microsoft.CognitiveServices/quotaTiers/default?api-version=2025-10-01-preview"

The response is a single object, not a value array. The following example response is abbreviated:

{
  "name": "default",
  "type": "Microsoft.CognitiveServices/quotaTiers",
  "id": "/subscriptions/00000000-0000-0000-0000-000000000000/providers/Microsoft.CognitiveServices/quotaTiers/default",
  "properties": {
    "assignmentDate": "2025-06-09T18:13:29.389Z",
    "currentTierName": "Free-Tier",
    "tierUpgradePolicy": "OnceUpgradeIsAvailable"
  }
}

Reference: Quota Tiers - Get.

Model-specific quota limits

For Azure OpenAI model limits by tier, see Azure OpenAI quota tier reference. For other Foundry Models, see the applicable model-specific quota sections in this article.

Check quota management scope

You can find the quota management system that applies to a given model by going to the Foundry portal's Quota page. The value in the Scope column for a given model indicates how Foundry manages quota for that model. A Scope value of:

• Global or Data Zone, indicates subscription-level quota management.
• A region (for example, East US or West US), indicates per-region quota management for that subscription and model.

Changes for onboarded models

For the models that are onboarded to the subscription-level quota management system:

• All Global Standard deployments of the same model and version under a subscription now draw from a single shared quota pool across all regions.
• All Data Zone Standard deployments of the same model and version under a subscription now draw from a shared quota pool within each data zone.
• Existing approved quota is retained and automatically applies at the subscription level—no action required.

This consolidation allows Microsoft Foundry to offer supported models consistently across all Foundry regions, regardless of how quota is distributed across resources or regions.

Quota limits when a model is upgraded

When an existing model is upgraded, its new Global or Data Zone quota limit is set to the greater of:

• The applicable tier limit.
• The total quota assigned to all existing deployments of that model within the quota scope.

For example, if a model has Global Standard deployments across five regions, the new global quota limit equals the combined quota of those deployments if that total exceeds the tier limit. Otherwise, the tier limit applies.

Grok 4.7 quotas

Grok 4.7 version 1 uses subscription-level quota management. The following table lists tokens per minute (TPM) quotas and corresponding requests per minute (RPM) limits by subscription tier. Each unit of deployment capacity provides 1,000 TPM and 1 RPM. Your deployment's rate limits depend on its configured capacity and allocated quota.

Microsoft-Decision-1 quotas

The following table lists tokens per minute (TPM) and requests per minute (RPM) limits for Microsoft-Decision-1 by subscription tier. Your deployment's rate limits depend on its configured capacity and allocated quota.

Quotas and limits reference

The following sections provide a quick guide to the default quotas and limits that apply to Foundry Models. Quotas and limits aren't enforced at the tenant level. Instead, the highest level of quota restrictions is scoped at the Azure subscription level. Tokens per minute (TPM) and requests per minute (RPM) limits are defined per region, per subscription, and per model or deployment type.

Resource limits (per Azure subscription, per region)

Rate limits

For Microsoft-Decision-1 rate limits, see Microsoft-Decision-1 quotas.

The following table lists limits for Foundry Models for the following rates:

• Tokens per minute
• Requests per minute
• Concurrent request

To increase your quota, use Microsoft Foundry Service: Request for Quota Increase to submit your request. Due to high demand, requests to increase quota are evaluated individually. For more information on quota increase requests, see request increases to the default limits.

Other limits

¹ Current APIs allow up to 10 custom headers, which the pipeline passes through and returns. If you exceed this header count, your request results in an HTTP 431 error. To resolve this error, reduce the header volume. Future API versions won't pass through custom headers. Don't depend on custom headers in future system architectures.

Usage tiers

Global Standard deployments use Azure's global infrastructure to dynamically route customer traffic to the data center with best availability for the customer's inference requests. This infrastructure enables more consistent latency for customers with low to medium levels of traffic. Customers with high sustained levels of usage might see more variabilities in response latency.

The Usage Limit determines the level of usage beyond which customers might see larger variability in response latency. A customer's usage is defined per model and is the total tokens consumed across all deployments in all subscriptions in all regions for a given tenant.

Request increases to the default limits

Submit the quota increase request form to request quota increases for Foundry Models sold by Azure, Azure OpenAI models, and Anthropic models. Except for Anthropic models, Models from partners and community don't support quota increases.

Quota increase requests are processed in the order they're received, and priority goes to customers who actively use their existing quota allocation. Requests that don't meet this condition might be denied.

General best practices to stay within rate limits

To minimize issues 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.

Setting client-side timeout

Set the client-side timeout explicitly based on the following guidance.

• Reasoning models (models that generate intermediate reasoning tokens before producing a summarized response): up to 29 minutes.
• Non-reasoning models:
  • For streaming, up to 60 seconds.
  • For non-streaming requests, up to 29 minutes.

29 minutes here doesn't mean all requests take 29 minutes but rather depending on context tokens, generated tokens, and cache hit rates, requests can take up to 29 minutes.

Set a timeout that's less than these values, tuned to your traffic patterns.

For reasoning models including streaming requests, all the reasoning tokens are first generated and then summarized before sending the first response token back to the user.

You can modify the reasoning effort parameter to control the number of reasoning tokens generated in the process.

Troubleshooting

Related content

• Models available in Foundry Models
• Manage and increase quotas for Foundry resources
• Quotas and limits in Azure OpenAI