This guide walks you through setting up AI Runway on an Azure Kubernetes Service cluster. You start with a bare cluster and end with a running AI model. The process includes checking your cluster, installing the controller, and assessing your GPU hardware.
You will also choose an inference provider like KAITO or Dynamo and deploy your first model. Each step is laid out in order. If you ever need to pick up from a specific step, you can use a skip command.
The guide also warns about GPU costs and helps you avoid common issues. It is built for anyone who wants to run large language models or other AI models on AKS.
Global
mkdir -p ~/.claude/skills/airunway-aks-setupProject
mkdir -p .claude/skills/airunway-aks-setupSource Repository
Azure Deploymicrosoft/azure-skills
Run Azure deployment commands with error recovery for already-prepared apps
Azure Preparemicrosoft/azure-skills
Generate Azure infrastructure code and config files for deployment preparation
Azure Storagemicrosoft/azure-skills
Store files, messages, and data with Azure Storage services easily.
Azure Validatemicrosoft/azure-skills
Validate your Azure deployment with thorough preflight checks on all prerequisites
Azure Resource Lookupmicrosoft/azure-skills
Find any Azure resource across subscriptions and groups easily
Azure Rbacmicrosoft/azure-skills
Find the least privilege Azure RBAC role and generate CLI and Bicep code
Azure Aigatewaymicrosoft/azure-skills
Govern your AI models and tools with Azure API Management gateway
Azure Resource Visualizermicrosoft/azure-skills
Map your Azure resources with clear architecture diagrams in minutes