customize

Guide Azure OpenAI model deployments with version, SKU, capacity, and RAI policy configuration.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/iinoshirozheng/SinoCloudChat --skill customize-iinoshirozheng
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: customize
Source: https://github.com/iinoshirozheng/SinoCloudChat/tree/main/.agents/skills/microsoft-foundry/models/deploy-model/customize
Command: npx skills add https://github.com/iinoshirozheng/SinoCloudChat --skill customize-iinoshirozheng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Deploying Azure OpenAI models with precise customization often requires juggling versions, SKUs, capacity, RAI policies, and advanced features across regions. This skill provides an interactive guided workflow to configure deployment settings end-to-end, ensuring you get the exact combination of model, capacity, and safeguards for production-grade use.

Core Features & Use Cases

  • Interactive, step-by-step deployment customization: choose model version, select SKU (GlobalStandard/Standard/ProvisionedManaged), set capacity (TPM or PTU), configure RAI policy, and enable advanced options like dynamic quota, priority processing, and spillover.
  • Cross-region planning and deployment orchestration: handles region availability, capacity checks, and deployment naming with auto-conflict resolution.
  • Anthropic support path and standard OpenAI deployment path: includes REST API deployment flow for non-OpenAI formats when required.
  • Use Case: You need a tightly controlled production deployment with guaranteed throughput and compliance checks, not just a quick regional install.

Quick Start

Provide the model name, desired version, and deployment options to begin the guided customization flow.

Frequently Asked Questions about customize

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

FAQPage Schema
How do I customize an Azure OpenAI deployment with specific SKU and capacity settings?▼

To customize an Azure OpenAI deployment, you select a model version, choose a SKU like GlobalStandard or ProvisionedManaged, and set capacity using TPM or PTU. The guided workflow enforces SKU availability and capacity ranges during creation.

What is a Provisioned Throughput Unit (PTU) setup for Azure OpenAI models?▼

A PTU setup for Azure OpenAI models allocates guaranteed throughput capacity instead of pay-per-use token billing. The deployment customization process supports PTU configurations to ensure production-grade performance with predictable capacity.

How do I configure RAI policy options during Azure OpenAI deployment?▼

RAI policy configuration is integrated into the Azure OpenAI deployment workflow. You can apply policy options and safeguards during the guided setup to ensure compliance checks are enforced before the model deployment goes live.

Can I use CLI and REST API to create customized Azure OpenAI deployments across regions?▼

Yes, Azure OpenAI customized deployments support both CLI and REST-based creation. The workflow handles cross-region capacity planning, checks region availability, and performs deployment naming with automatic conflict resolution.

What advanced options are available when deploying Azure OpenAI models for production?▼

Advanced options for Azure OpenAI deployments include dynamic quota, priority processing, and spillover configurations. These features are configured alongside SKU selection and capacity settings to meet strict production requirements.

When should I use ProvisionedManaged versus Standard SKU for Azure OpenAI deployment?▼

Use ProvisionedManaged SKU for guaranteed throughput with PTU capacity, and Standard SKU for TPM-based flexible scaling. The deployment workflow enforces both SKU availability and capacity range constraints based on your chosen configuration.