API gateway cost compared: API Gateway vs API Management vs Cloud endpoints

Quick answer

AWS API Gateway bills per request (HTTP APIs cheaper than REST), scaling to zero. Azure API Management has a Consumption tier (per call, scales to zero) and provisioned tiers (fixed baseline). GCP offers a serverless API Gateway (per call) and Cloud Endpoints/Apigee for more. The cheapest depends on volume and features: per-request/serverless options win for variable traffic and basic needs; provisioned tiers win at high steady volume with heavy feature use.

API management across the three clouds ranges from lightweight per-request gateways to full platforms with developer portals and monetization. The pricing spans per-request, per-call, and tiered subscription models, so the cheapest depends on your traffic volume and how many management features you actually need.

The options and their pricing

Cloud	Lightweight (per request)	Full platform (tiered)
AWS	API Gateway (HTTP/REST, per request)	REST API with usage plans
Azure	API Management Consumption (per call)	API Management Standard/Premium
GCP	API Gateway / Cloud Endpoints (per call)	Apigee (subscription)

AWS API Gateway is per request, with HTTP APIs cheaper than REST APIs, as in REST versus HTTP API. Azure API Management offers a serverless Consumption tier (per call, scales to zero) and provisioned Standard/Premium tiers with a fixed baseline, as in API Management cost. GCP has a lightweight serverless API Gateway and Cloud Endpoints, with Apigee for full platform needs.

Volume and features decide

For variable or low traffic and basic gateway needs (routing, auth, rate limiting), the per-request and serverless options (AWS API Gateway, Azure APIM Consumption, GCP API Gateway) are cheapest because they scale to zero. For high, steady traffic with heavy feature use, provisioned tiers can be cheaper per call and add features (caching, developer portals). For a full API-as-a-product program with monetization, the platform options (Apigee, APIM Premium) are the tool despite their baseline cost.

The shared truth: backend costs more

On all three, the API gateway usually costs less than the backend each request invokes, the Lambda, function, container, or database behind it. So while the gateway comparison matters, optimizing the backend, and caching at the gateway where supported to deflect backend calls, often saves more than the gateway choice itself.

Choosing on cost

Use the gateway on the cloud where your backend runs, choose a per-request/serverless option for variable traffic and basic needs, a provisioned tier for high steady traffic or when its features justify the baseline, and a full platform only for a genuine API product. Cache at the gateway to deflect backend cost. The gateway rates are close enough that features and volume, plus backend optimization, matter more than the provider.

FAQ

Is AWS API Gateway, Azure API Management, or GCP API Gateway cheapest?

It depends on volume and features. Per-request and serverless options (AWS API Gateway, Azure APIM Consumption, GCP API Gateway) are cheapest for variable traffic and basic needs since they scale to zero. Provisioned tiers can be cheaper at high steady volume with heavy feature use. The cheapest is usually the gateway on your backend's cloud.

How do API gateway pricing models compare across clouds?

AWS API Gateway bills per request (HTTP cheaper than REST). Azure API Management has a Consumption tier (per call, scales to zero) and provisioned Standard/Premium tiers with a fixed baseline. GCP offers a serverless API Gateway (per call) plus Apigee for full platform needs. Models span per-request, per-call, and subscription.

When should I use a per-request versus a provisioned API gateway?

Use a per-request or serverless option for variable or low traffic and basic gateway needs (routing, auth, rate limiting), since it scales to zero. Use a provisioned tier for high, steady traffic where it is cheaper per call and adds features, and a full platform like Apigee only for a genuine API-as-a-product program.

What is the real cost of an API gateway?

Usually the backend each request invokes, the Lambda, function, container, or database behind the gateway, costs more than the gateway itself. So while the gateway comparison matters, optimizing the backend and caching at the gateway to deflect backend calls often saves more than the gateway choice.

Should I choose a cloud based on API gateway cost?

No. The gateway rates are close, so choose the gateway on the cloud where your backend runs, select the tier that matches your traffic and features, and cache at the gateway to deflect backend cost. Backend optimization and feature-to-need matching matter more than the marginal provider difference.

Does C3X estimate API gateway cost?

C3X prices API gateway resources across clouds from your Terraform, including provisioned tiers. Per-request and per-call cost depends on traffic volume, which you model as a usage assumption, along with the backend services the gateway fronts, where much of the cost sits.

What to do next

Compare API management options for your workload. C3X reads your Terraform and prices your resources against a live catalog. Start with the quickstart.