Share via

Claude Opus Capacity Deniel in all regions - Sweden, East US2

Sudhanshu Shekhawat 0 Reputation points Microsoft Employee
2026-04-30T06:30:48.3433333+00:00

Hi Team,

We have been trying to request Opus models quota for different subscription, and it is very less than 15000 tokens per minute. But it gets denied, even for low-capacity request. 

Subscriptions:

<Sub and Tenant id redacted at support side>

Requirements:

  1. Our service uses total token of around 60k tokens every minute ( given 2 request comes every minute)
  2. For testing, it is fine to give 15000 tokens - 30000 tokens per minute.  Subscriptions: <Sub and Tenant id redacted at support side>    Requirements:
    1. Our service uses total token of around 60k tokens every minute ( given 2 request comes every minute)
    2. For testing, it is fine to give 15000 tokens - 30000 tokens per minute. 

Can you please help approve the request or suggest what should be done to move forward?

Note : Sub and Tenant id redacted at support side for confidentiality of customer

Foundry Models
Foundry Models

A catalog of AI models in Microsoft Foundry that you can discover, compare, and deploy using Azure’s built‑in tools for evaluation, fine‑tuning, and inference

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Manas Mohanty 16,670 Reputation points Microsoft External Staff Moderator
    2026-05-04T08:50:24.8266667+00:00

    Hi Sudhanshu Shekhawat/Bhavesh

    Got inputs from SME from internal channel.

    There is restriction on Azure Native usage of Claude model for 1P customers.

    But they can use Claude Direct usage (with custom function/OpenAPI probably) with approval from Microsoft CELA

    Please contact you respective Microsoft CELA team for the same.

    Relevant document outlining restriction has been passed to Sudhansu for reference.

    Thank you for understanding our constraints on support side on the same.

    0 comments

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.