Managed containers (no orchestration) · Container-only

Azure Container Instances

Run a single container or container group without provisioning VMs or orchestration.

When to use it

Choose Container Instances for simple, isolated container execution — a single container or container group, batch tasks, build agents, or short-lived jobs — where you don't need orchestration, scaling, or load balancing. It starts fast and bills per second.

When not to use it

Avoid Container Instances when you need autoscaling, load balancing, rolling updates, or to run many containers as a service — those need Container Apps or AKS. With no built-in autoscaling or load balancer, it isn't suited to production web traffic on its own.

Trade-offs at a glance

Hosting model Managed containers (no orchestration)
Container support Container-only
Minimum nodes No dedicated nodes
State management Stateless
Web hosting Agnostic
Autoscaling Not supported
Load balancer No built-in support
Scale limit 100 container groups per subscription (default)
Multiregion Single region only
Virtual network integration Supported
Hybrid connectivity Supported
GPU support Not supported
TLS Use a sidecar container
Architecture styles Microservices, task automation, batch jobs
Required skills Container basics
Operational overhead Very low — no orchestration
Best for teams Teams that need simple container execution

Frequently asked questions

Does Container Instances support autoscaling?

No. Container Instances has no built-in autoscaling or load balancing. For scaling containers, use Container Apps or AKS; ACI is best for single or fixed-count container workloads.

What is a container group?

A container group is a set of containers scheduled on the same host that share lifecycle, network, and storage — similar to a Kubernetes pod. It's the deployment unit for Linux in ACI.

Can I use Container Instances for scheduled jobs?

Yes. ACI suits task automation and batch jobs — start a container group, run the task, and it stops when complete, billing only for the runtime.

Starting point, not a verdict

This is one candidate from the Azure compute decision guide. The right choice depends on your full requirements — evaluate scaling, cost, and operational fit before committing.

Browse all tools →