PA - How to Request Azure OpenAI Quotas

PA - How to Request Azure OpenAI Quotas

Introduction

Quota increase requests are submitted via this request form. Please submit one request for each of the following three models:

  • GPT-4o

  • GPT-4o-mini

  • DALL-E

Form Questions

The form questions are numbered and the number is given at he end of each section title.

User and Company Information (1-8)

The first few questions pertain to your personal and company information:

  • First Name

  • Last Name

  • Company Email

  • Company Name

  • Company Address

  • Company City

  • Company Postal Code

  • Company Country

Subscription ID (9)

The Azure Subscription ID can be found in the Azure portal. Use the search bar to locate Subscriptions and select the one you are using. The Subscription ID will be listed in the overview section.

Justification (10)

Please enter the following justification:

The Ayfie Personal Assistant application that my company uses requires a quota increase due to its reliance on OpenAI models to process large documents. The application utilizes language models to generate detailed answers, which can consume a significant number of tokens, especially with extensive documents. Increasing the quota is necessary for the application to handle larger datasets efficiently and provide timely responses.

Quota Request Type (11)

Select Standard as the Quota Request Type.

Standard Region (12)

The preferred region for deploying the Personal Assistant application is Sweden Central, as this region offers the highest availability of models within the European Union. If you are located in the US, then select East-US instead.

Region Availability (13)

Select the Grant me a quota in an alternate region option.

Alternate Region (14)

Select Anywhere in the EU or Anywhere in the USA depending on your company’s location.

Standard Model (15) and Standard Quota (16)

You must complete the form three times, once for each of the model listed below along with the quota to request:

  • GPT-4o: 150 capacity units

  • GPT-4o-mini: 450 capacity units

  • DALL-E: 2 capacity units