API gateway.
Managed gateway with auth, rate limiting, transformation and usage plans for your APIs.
Front your services with a managed gateway that handles auth, rate limiting, transformation and usage plans — so each backend doesn't reinvent the edge, and partners get a clean, metered API.
- Category
- Integration
- Deployment
- Managed → air-gapped
- Governance
- IAM · encryption · audit
Every backend reinvents auth, limits and metering. The gateway is where that stops.
Without a gateway, each service grows its own copy of API keys, rate limiting, request validation and usage tracking — inconsistently, and usually badly. A managed gateway moves that to the edge so backends do one thing. Aether's gateway handles auth (keys, OAuth, JWT), per-key and per-route rate limits, request/response transformation and metered usage plans — and fronts your model endpoints the same way it fronts any service.
One front door for services and the model alike.
The distinctive part: your AI endpoints are APIs too, and they need exactly the same edge concerns — auth, quotas, transformation, monetisation. Putting model endpoints behind the same gateway means a partner API and an inference API are governed, rate-limited and metered identically, with usage plans you can turn into a product.
Versioned, observable, and on the platform.
APIs are versioned so you can evolve without breaking consumers, every call flows into the same observability as the rest of the stack, and policy is enforced consistently. It deploys in your boundary, so the gateway and the services behind it share one identity and one audit trail.
Put a model endpoint behind the gateway with auth, a per-key quota and a usage plan — the same way you'd expose any API.
route: /v1/summarize
upstream: aether.serving/aether-40b
auth: jwt
rate_limit: { per_key: 100/min }
plan: { tier: pro, meter: tokens } # metered, monetisable
transform: { strip: [internal_id] }The model endpoint gets auth, quotas, metering and a usage plan from the same gateway that fronts your other APIs — no bespoke wrapper around inference. A partner API and an AI API are governed identically.
Three steps to running.
Put the gateway in front of any backend with auth and rate limits.
Transform requests and responses; meter with usage plans.
Expose clean, versioned APIs to partners and consumers.
API gateway, in full.
API keys, OAuth and JWT validation at the edge.
Per-key and per-route quotas with burst control.
Reshape requests and responses without touching the backend.
Meter and monetize APIs with per-consumer tiers.
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 api gateway on Aether Cloud
aether integration create \
--service api-gateway \
--name app \
--region us-1 \
--deploy managed # or vpc | on-prem | air-gappedAt a glance.
- Auth
- Keys · OAuth · JWT
- Limits
- Per-key and per-route
- Transform
- Request/response
- Plans
- Metered usage tiers
- Versioning
- Built-in
Built for real work.
Public and partner APIs
Microservice fronting
Monetized API products
On one model, not stitched together.
The usual stack runs api gateway 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 — api gateway sits next to the rest of the catalog, one identity, one bill.
The provider that runs Aether runs your api gateway — 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.
A managed gateway with auth, limits and transformation — fronting your services and AI endpoints alike.
Yes — usage plans meter and tier per consumer.
Yes — API keys, OAuth and JWT validation at the edge.
Your boundary, your choice.
Pairs well with.
Durable queues for decoupled, at-least-once processing with dead-letter handling.
Publish-subscribe event routing with filtering and fan-out across services.
Durable, stateful workflows for long-running processes, retries and agent pipelines.
mTLS, traffic management and observability between your services, without app changes.
Run API gateway on Aether Cloud.
Managed gateway with auth, rate limiting, transformation and usage plans for your APIs. Deployable managed, in your VPC, on-prem or fully air-gapped — talk to us about the configuration your workloads and your boundary require.