preset

Deploy Azure OpenAI models to regions with real-time capacity via Azure REST.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/attentiondotnet/azure-skills --skill preset-attentiondotnet
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: preset
Source: https://github.com/attentiondotnet/azure-skills/tree/main/.github/plugins/azure-skills/skills/microsoft-foundry/models/deploy-model/preset
Command: npx skills add https://github.com/attentiondotnet/azure-skills --skill preset-attentiondotnet

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Deploying Azure OpenAI models often fails because the selected region lacks capacity or the project does not exist, requiring manual checks, quota requests, and region selection. This Skill automates those decisions so the model reaches a live endpoint without user‑driven capacity hunting.

Core Features & Use Cases

  • Capacity Awareness: Queries real‑time model capacity in the current region and all Azure locations, falling back to the best available region automatically.
  • Region & Project Management: Detects missing projects, creates new AI Services hubs and Foundry projects in the chosen region, and resolves naming conflicts.
  • Anthropic Support: Handles Anthropic models via ARM REST calls with required industry data prompts.
  • Error Recovery: Gracefully handles quota exhaustion, suggests quota increase, and offers alternative models.
  • Use Case Example: An engineer can request a gpt-4o deployment without knowing which Azure region has capacity; the Skill selects East US 2, creates any missing project, and completes the deployment in minutes.

Quick Start

Deploy the gpt-4o model using the preset skill, letting it automatically choose the best region and handle any capacity issues.

Frequently Asked Questions about preset

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy Azure OpenAI models to a region with available capacity?

To deploy Azure OpenAI models to a region with available capacity, you can automatically query real-time model capacity across all Azure locations and fall back to the best available region. This handles quota limits and region selection without manual capacity hunting.

What happens if my Azure OpenAI deployment fails due to quota exhaustion?

When Azure OpenAI deployment fails due to quota exhaustion, the process gracefully handles the error by suggesting a quota increase and offering alternative models. It also automatically checks other Azure locations for available capacity to complete the deployment.

Does Azure AI Foundry require an existing project before model deployment?

Azure AI Foundry does not require an existing project before model deployment. The deployment process detects missing projects and automatically creates new AI Services hubs and Foundry projects in the chosen region, resolving any naming conflicts.

Can I deploy Anthropic models on Azure using ARM REST calls?

Yes, you can deploy Anthropic models on Azure using ARM REST calls. The deployment process handles Anthropic models by submitting the required industry data prompts through ARM REST to provision the resources correctly.

What is the best way to automate Azure OpenAI region selection for provisioning?

The best way to automate Azure OpenAI region selection is to query real-time capacity via Azure REST and automatically select the optimal region. This approach manages automated authentication and creates missing AI projects to ensure a live endpoint.