Announcing Global Provisioned Managed Deployments for Scaling Azure OpenAI Service Workloads

We are excited to announce significant progress in AI deployment. Azure OpenAI Service: Global provisioning managed deployment is generally available (GA) starting September 18, 2024. This launch represents a significant milestone in our efforts to make AI more accessible, scalable, and flexible for customers around the world, and builds on the provisioning throughput unit (PTU) launch in August. Self-service area distribution.

What is Global Provisioning Management?

Global Provisioned Managed is a new deployment type within the Azure OpenAI service that leverages Azure’s global infrastructure to provide provisioned traffic more efficiently. It supports the latest GPT-4o (2024-08-06) and GPT-4o-mini (2024-07-18) models, allowing customers to access them without regional quotas or capacity restrictions. This new deployment model gives customers greater flexibility and speed in deploying models by expanding their AI capabilities to anywhere in the world.

Dual availability: global and regional

We are also excited to announce that the GPT-4o (2024-08-06) model is now available for provisioning regional deployments via self-service in addition to global provisioning managed deployments. This means that customers have the flexibility to choose between a global managed deployment model or a more controlled regional deployment approach, depending on their specific needs and preferences.

Key Benefits of Global Provisioning Management Deployment

1. Access to the latest models is available everywhere: The global provisioning managed deployment model removes regional restrictions, allowing customers to access the latest AI models such as GPT-4o and GPT-4o-mini in all supported Azure regions, including eastus, westeurope, and japaneast.
2. Simplified deployment and management: Unlike traditional deployment methods, Global Provisioned Managed isolates capacity management for specific regions, automatically granting all eligible customers access to new global quotas.
3. Data Residency and Compliance Flexibility: API traffic may be processed globally, but all customer data is securely stored in the region of the Azure OpenAI service resource to ensure compliance. Local Data Residency and Compliance Requirements.
4. Transparent and flexible pricing: Global provisioning managed billing follows the same model as traditional provisioning managed deployments, ensuring predictable costs with hourly pricing and reservation options to suit a variety of usage scenarios.
5. Dual deployment options for greater flexibility: The GPT-4o model is available for both global provisioning managed deployments and provisioning regional deployments, giving customers the freedom to choose the deployment strategy that best suits their organization’s needs.

Why choose Global Provisioned Managed?

This new deployment type represents a significant evolution in our approach to AI, offering:

• Global Reach: Deploy AI models anywhere without regional quota or capacity constraints.
• Cost-effectiveness: Take advantage of cost management options, including monthly and annual bookings.
• Enhanced flexibility: Reduce complexity and management burden so you can deploy and scale AI solutions faster, freeing you to focus more on innovation.
• Local control: For customers requiring regional deployments, the GPT-4o model remains available via self-service, giving you full control over capacity management.

How to get started

Deploying AI models globally or locally is simple.

• For global provisioning management deployments: This option will be available for Azure OpenAI service regional resources starting September 18, 2024. To use it, create a regional resource or select an existing one and choose the Global provisioning managed deployment option.
• For provisioned regional deployments: that GPT-4o(2024-08-06) This model can be used for self-service regional deployments, allowing you to flexibly manage regional capacity and resources based on your users’ needs.

Looking ahead: More models and regions

The initial release of Global Provisioned Managed includes support for GPT-4o and GPT-4o-mini models, with plans to expand availability of more models in this deployment type. If you require specific regional support, you can continue to use your existing Provisioned Managed deployments.

Embrace the future of AI with Azure OpenAI services

The Azure OpenAI service is dedicated to pushing the boundaries of AI capabilities. With the new Global Provisioned Managed deployment, we are breaking down barriers, providing more flexibility, and ensuring that customers can unleash the full potential of AI anywhere in the world.

Learn more: