Azure API Management Consumption Tier - Actual operations limit and rollout status

KUENTZ Nicolas 20 Reputation points
2026-08-17T12:01:59.1266667+00:00

Hello,

I am trying to better understand the current API Operations limit for Azure API Management Consumption tier and whether the 2026 limits rollout has been fully completed.

According to the current documentation, the Consumption tier should support up to 3,000 API operations.

My observations are the following:

Environment:

  • Tier: Consumption
  • Region: France Central
  • New APIM instance created recently

Testing performed:

  1. Imported 999 operations -> Success
  2. Imported another 999 operations -> Success
  3. Total operations in the service: 1,998
  4. Attempted to import an additional 999 operations -> Failed

The error returned is:

"You've reached the maximum number of Operations (1000) in Service. Please delete one or more Operation(s) from Service to continue, or consider upgrading your plan to increase the limit."

What is confusing is that the service already contains 1,998 operations, which suggests that the effective limit is not actually 1,000 operations anymore.

Could Microsoft clarify:

  • What is the current effective API operations limit for Consumption tier in France Central?
  • Has the 3,000 operations rollout been completed?
  • Is the error message outdated or misleading?
  • Are there any known differences between regions, stamps, or subscriptions?

Thank you.

Azure API Management
Azure API Management

An Azure service that provides a hybrid, multi-cloud management platform for APIs.

0 comments No comments

Answer accepted by question author
Allan Solomon Mejia 6,240 Reputation points
2026-08-17T19:14:32.3066667+00:00

Hello @KUENTZ Nicolas

According to the current Microsoft documentation, the Azure API Management Consumption tier supports up to 3,000 API operations per service instance. Microsoft announced that the updated limits would begin rolling out to Consumption instances from March 15, 2026.

Your result is therefore unusual. If the service already contains 1,998 operations, but the next import fails with:

"You've reached the maximum number of Operations (1000) in Service"

then the behavior doesn't match the currently documented 3,000-operation limit.

One thing worth noting is that APIM counts API-related resources across the service, including API versions and revisions, so I would first verify the actual operation count through the REST API/ARM rather than relying only on what is visible in the portal.

However, even so, the explicit 1,000-operation limit in the error message suggests the instance may still be subject to older limit enforcement, or that there is an inconsistency between the documented limit and the backend enforcement for that service/stamp.

Since this is a newly created Consumption instance in France Central, I'd recommend opening an Azure support request and providing the APIM service name, region, subscription ID, timestamp of the failed deployment, correlation/request ID, and the complete error response. Microsoft Support can confirm the effective backend quota assigned to that specific instance.

The currently published limit is 3,000 operations for Consumption, so a newly created instance stopping around 2,000 operations with a 1,000-operation error warrants investigation by the APIM product team.

Sharing these references with you:

Microsoft Learn: Azure subscription and service limits, quotas, and constraints | https://learn.microsoft.com/en-us/azure/azure-resource-manager/management/azure-subscription-service-limits?

Microsoft: New Azure API Management service limits | https://learn.microsoft.com/mt-mt/azure/api-management/service-limits?

Please "Accept the Answer" if this information helped you. This will help us and others in the community as well.

Was this answer helpful?

1 person found this answer helpful.
0 comments No comments

0 additional answers

Sort by: Most 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.