Containers.
Managed container runtime with autoscaling, health checks and rolling deploys — bring an image, get a URL.
Run containers without operating a cluster. Push an image and the platform handles scheduling, autoscaling, health checks and zero-downtime rollouts behind a managed URL.
- Category
- Compute
- Deployment
- Managed → air-gapped
- Governance
- IAM · encryption · audit
Image-to-URL is solved. The seams to the model and data aren't.
Push an image, get an autoscaling service behind a URL — Cloud Run and Fargate proved how good that experience can be, and Aether matches it: scheduling, autoscaling to zero, rolling and blue-green deploys, health gates and instant rollback, no cluster to operate. What changes the architecture is what the container sits next to. Here it's the model, the vector store and your governed data, wired in by default rather than reached across a vendor boundary.
Autoscaling and safe deploys without a cluster.
Scale on CPU, memory, concurrency or a custom metric, down to zero when idle. Roll out with health checks and revert in one click. Secrets, private networking and observability attach automatically, so a service is production-shaped from the first deploy, not after a hardening pass.
When you outgrow it, Kubernetes is right there.
This is the no-cluster path for stateless services; the moment you need the full ecosystem and control, the managed Kubernetes in the same catalog is one step away, on the same platform with the same identity. It all deploys managed or air-gapped.
Deploy a container to a URL with autoscaling and the model reachable in-process — no cluster, no external AI hop.
aether containers deploy \
--image registry.aether/api:latest \
--scale "0..50 on concurrency" \
--secret DB_URL --observe
# -> https://api.app.aether — model + vector store reachable locallyThe service scales to zero and back, and the code inside reaches the model and data on the same platform. On Fargate or Cloud Run those are external calls across a boundary you have to secure.
Three steps to running.
From the registry or any OCI image; the platform schedules and routes it behind a URL.
Scale on CPU, memory, concurrency or a custom metric — down to zero when idle.
Rolling or blue-green rollouts with health gates and instant rollback.
Containers, in full.
Deploy from any registry; the platform provisions, routes and scales it for you.
Scale on CPU, memory, concurrency or a custom metric — down to zero when idle.
Zero-downtime deploys with health gates and instant rollback.
Secrets, private networking and observability attached automatically.
Provision it in a few lines.
Every service is reachable from the same SDK, CLI and infrastructure-as-code — one identity, one bill, one audit trail across the whole catalog.
# Provision containers on Aether Cloud
aether compute create \
--service containers \
--name app \
--region us-1 \
--deploy managed # or vpc | on-prem | air-gappedAt a glance.
- Image source
- Any OCI registry
- Scaling
- Metric-based, to zero
- Rollouts
- Rolling · blue-green
- Networking
- Private VPC + managed URL
- Wired in
- Secrets · observability
Built for real work.
Stateless web and API services
Background workers
Packaged third-party software
On one model, not stitched together.
The usual stack runs containers in one product, the model in another and the data in a third — and the seams between them are the cost. Aether Cloud runs it on the same platform that serves the model, governs your identity and deploys into your boundary, with the rest of the catalog one hop away.
No stitching a vector DB to one place, a warehouse to another and a model to a third — containers sits next to the rest of the catalog, one identity, one bill.
The provider that runs Aether runs your containers — so the data and the model never leave the same governed boundary to talk to each other.
Need a capability that isn’t here yet? The model writes and deploys it into the same boundary — the catalog is a starting point, not a ceiling.
Good to know.
Same image-to-URL, serverless-container model — on the platform that runs the model and your data services next door.
No — this is the no-cluster path. Reach for Managed Kubernetes when you need the full ecosystem and control.
Best for stateless services; attach block or object storage for state where needed.
Your boundary, your choice.
Pairs well with.
On-demand and reserved VMs across CPU and GPU shapes, with per-second billing and live resize.
Latest-generation accelerators by the hour or the cluster, with fast interconnect for training and high-throughput inference.
Event-driven functions that scale to zero — run code and agents without managing a server.
A conformant, managed control plane with GPU scheduling, autoscaling node pools and zero-downtime upgrades.
Massively parallel batch and simulation jobs with scheduling, checkpointing and fault tolerance built in.
Dedicated single-tenant hardware for the workloads that need full control of the silicon.
Run Containers on Aether Cloud.
Managed container runtime with autoscaling, health checks and rolling deploys — bring an image, get a URL. Deployable managed, in your VPC, on-prem or fully air-gapped — talk to us about the configuration your workloads and your boundary require.