Intel® AI for Enterprise Solutions Documentation

The technical documentation for deploying and operating the platform. Start with Get started to understand what it is and stand it up, use the Deploy guides to take it further, then configure and look things up as needed.

Looking for the three-step quick start? It's in the project README.

Get started

Understand what the platform is, check your machine meets the bar, then deploy it.

Guide	What it covers
Meet Intel® AI for Enterprise Solutions	What the platform is, the problem it solves, and the layers it deploys
Prerequisites	Everything needed before running the installer
Getting Started	Three deployment paths — pick one and follow it end to end

Deploy

Install the platform, serve a model, and go beyond a single node.

Guide	What it covers
Deployment Guide	Step-by-step platform install — no Kubernetes experience needed
Deploy a Model	Deploying, accessing, and managing LLM inference with `model-manager`
Multi-Node & BYO Cluster	Deploying across several machines, or onto an existing cluster
Network Architecture	Network topology, IP allocation, and how a request reaches a model

Configure & customize

Tune the deployment: change any setting, place workloads across nodes, and connect your own applications.

Guide	What it covers
Configuration Reference	Every option in `global_config.yaml`, and how to override them
Integration Guide	Connecting your tool or framework without modifying the stack
Node Topology & Workload Placement	Separating platform and inference workloads on multi-node clusters
NRI CPU Balloons	NUMA-aware CPU pinning for inference pods

Reference

Look up exact commands, architecture decisions, and cluster-level details.

Guide	What it covers
CLI Reference	`es_auto_installer.sh` and `model-manager` commands and flags
Architecture & Design	How the modular deployment framework is structured
Namespace Security Labels	Pod Security Admission and Istio labels applied per namespace