WIP

Author: RoseHJM

articles/dev-box/concept-dev-box-serverless-gpu.md

@@ -1,8 +1,8 @@
 ---
-title: Serverless GPU compute in Microsoft Dev Box
+title: Serverless GPU Compute in Microsoft Dev Box
 description: Learn about serverless GPU compute in Microsoft Dev Box, how it works, benefits for developers and organizations, and key use cases.
 ms.service: dev-box
-ms.topic: concept-article
+ms.topic: how-to
 ms.date: 05/05/2025
 author: RoseHJM
 ms.author: rosemalcolm
@@ -11,9 +11,9 @@ ai-usage: ai-generated
 #customer intent: As a business decision-maker, I want to evaluate serverless GPU compute in Dev Box so that I can determine its value for my team’s workflows.
 ---
 
-# Serverless GPU compute in Microsoft Dev Box
+# Use Serverless GPU compute in Microsoft Dev Box
 
-Serverless GPU compute in Microsoft Dev Box lets you use GPU resources on demand for compute-intensive tasks like AI model training, inference, and data processing. Microsoft Dev Box serverless GPU compute enables developers to access powerful GPU resources on demand without requiring permanent infrastructure provisioning or complex setup. This article explains what serverless GPU compute is, how it works, and key scenarios for its use.
+Serverless GPU compute in Microsoft Dev Box (preview) lets you spin up dev boxes with GPU acceleration — no additional setup required. Microsoft Dev Box serverless GPU compute enables developers to access powerful GPU resources on demand without requiring permanent infrastructure provisioning or complex setup. This article explains what serverless GPU compute is, how it works, and key scenarios for its use.
 
 ## What is serverless GPU compute?
 
@@ -44,16 +44,16 @@ Serverless GPU compute in Microsoft Dev Box offers distinct benefits for both de
 
 - **No setup required**: Access GPU compute with a single click from your Dev Box environment
 - Access GPU compute with one step from your Dev Box environment.
-- **No permission barriers**: Use GPU resources without needing rights to create cloud infrastructure
-- **Integrated development experience**: Seamlessly use GPU compute within familiar tools like Windows Terminal, Visual Studio, and VS Code
-- **Zero configuration**: GPU sessions start automatically when needed and stop when not in use
+- **No permission barriers**: Use GPU resources without needing rights to create cloud infrastructure.
+- **Integrated development experience**: Seamlessly use GPU compute within familiar tools like Windows Terminal, Visual Studio, and VS Code.
+- **Zero configuration**: GPU sessions start automatically when needed and stop when not in use.
 
 ### For organizations
 
 - **Cost optimization**: Pay only for the GPU resources you use instead of provisioning dedicated hardware.
-- **Centralized control**: Manage GPU access through project-level policies
-- **Security and compliance**: Keep sensitive data in your secure corporate network while using GPU resources
-- **Simplified resource management**: Control GPU usage limits at the project level
+- **Centralized control**: Manage GPU access through project-level policies.
+- **Security and compliance**: Keep sensitive data in your secure corporate network while using GPU resources.
+- **Simplified resource management**: Control GPU usage limits at the project level.
 
 ## How serverless GPU compute works
 
@@ -69,47 +69,43 @@ Serverless GPU compute in Dev Box uses Azure Container Apps (ACA) to provide GPU
 These GPU options are supported:
 
 - **NVIDIA T4 GPUs**: Readily available with minimal quota concerns
-- **NVIDIA A100 GPUs**: More powerful but available in limited capacity
+
 
 ### Regional availability
 
 GPU resources are available in these Azure regions:
 
 - West US 3
-- Sweden North
-- Australia East
 
 Additional regions may be supported in the future based on demand.
 
-### vNet injection
-
-vNet injection allows customers to integrate their network and security protocols with the serverless GPU environment. While not required for the proof of concept (POC), this feature will be prioritized for public previews and general availability (GA). With vNet injection, customers can achieve tighter control over network and security configurations.
+## Administration and management
 
-vNet injection lets customers control network and security configurations more tightly.
+Admins control serverless GPU access at the project level through Dev Center. Key management capabilities include:
 
-### MOBO architecture model
+- **Enable/disable GPU access**: Control whether projects can use serverless GPU resources
+- **Set concurrent GPU limits**: Specify the maximum number of GPUs that can be used simultaneously across a project
 
-Serverless GPU compute adopts the MOBO architecture model for ACA integration. In this model, ACA instances are created and managed within the customer’s subscription, providing a more controlled and streamlined management experience. This ensures that the Dev Box service can securely manage ACA sessions without introducing additional complexity.
+Access to serverless GPU resources is managed through project-level properties. When the serverless GPU feature is enabled for a project, all Dev Boxes within that project automatically gain access to GPU compute. This simplifies the access model by removing the need for custom roles or pool-based configurations.
 
-### Developer experience
+## Configure Serverless GPU
 
-Developers can access serverless GPU compute through:
+To configure 
 
-- **Windows Terminal**: Launch a GPU-powered shell directly from Windows Terminal
-- **Visual Studio**: Access GPU compute from within the Visual Studio environment
-- **VS Code with AI Toolkit**: Use seamless GPU integration for AI development tasks
+### Register for the subscription
 
-The goal is to provide a seamless, native experience where GPU resources are accessible without requiring any setup from the developer.
+1. Sign in to the [Azure portal](https://portal.azure.com).
+1. Navigate to your subscription. 
+1. Select **Settings** > **Preview features**.
+1. Select **Dev Box Serverless GPU Preview**, and then select **Register**.
 
-## Administration and management
+### Enable GPU for a project
 
-Admins control serverless GPU access at the project level through Dev Center. Key management capabilities include:
+1. Go to your project.
+1. Select **Settings** > **Dev box settings**.
+1. Under **AI workloads**, select **Enable**, and then select **Apply**.
 
-- **Enable/disable GPU access**: Control whether projects can use serverless GPU resources
-- **Set concurrent GPU limits**: Specify the maximum number of GPUs that can be used simultaneously across a project
-- **Cost controls**: Manage GPU usage within subscription quotas
 
-Access to serverless GPU resources is managed through project-level properties. When the serverless GPU feature is enabled for a project, all Dev Boxes within that project automatically gain access to GPU compute. This simplifies the access model by removing the need for custom roles or pool-based configurations.
 
 ## Related content