Home/Library/API Gateway REST vs HTTP Cost
Comparison · AWS Cost Optimization · Updated June 2026

Amazon API Gateway Cost: REST vs HTTP APIs Compared

HTTP APIs cost roughly a third of REST APIs for the same traffic. The price gap is large enough that the real question is rarely which is cheaper, but whether you use any of the REST-only features that justify paying more than triple the per-request rate.

Key takeaways

Amazon API Gateway HTTP APIs are about 71 percent cheaper than REST APIs: roughly $1.00 per million requests versus $3.50 in US East. Choose an HTTP API for the common pattern of routing to Lambda or an HTTP backend with JWT auth, CORS, and throttling. Choose a REST API only when you need response caching, usage plans and API keys, request validation, AWS WAF, or private VPC endpoints. The biggest hidden REST cost is per-hour caching, billed whether or not it is hit. Default to HTTP and upgrade a single route to REST only where a feature demands it.

Last updated: June 2026 · Written by Morten Andersen and reviewed by Fredrik Filipsson · See, Cut, Lock, Run method

Amazon API Gateway is a managed service for publishing, securing, and operating APIs, and it offers two request-priced flavors that do largely the same job at very different prices. An HTTP API is the newer, leaner option built for low-latency proxying to Lambda and HTTP backends; a REST API is the older, fuller option with a deep feature set for API management. The cost difference is not subtle: HTTP APIs are billed at roughly a third of the REST rate, so for most serverless backends the cheaper tier is also the right tier. This article is part of our complete guide to AWS cost optimization, the cluster pillar it links up to.

How much cheaper are HTTP APIs than REST APIs?

HTTP APIs are roughly 71 percent cheaper than REST APIs on a per-request basis. In US East (N. Virginia), HTTP APIs are billed at about $1.00 per million requests for the first 300 million calls per month, while REST APIs are about $3.50 per million for the first 333 million, with both stepping down at higher volumes. That gap means a service handling a billion calls a month pays on the order of $1,000 on HTTP APIs versus about $3,500 on REST for the request charge alone. Prices vary by Region and AWS adjusts them over time, so confirm the current numbers on the Amazon API Gateway pricing page before modeling a migration.

CriterionHTTP APIREST API
Price per million (US East, first tier)~$1.00~$3.50
Lambda & HTTP proxy integrationYesYes
JWT auth, CORS, custom domains, throttlingYesYes
Response cachingNoYes (per-hour charge)
Usage plans & API keysNoYes
Request/response validation & transformationLimitedFull
AWS WAF integrationNoYes
Private APIs (VPC endpoint)NoYes

The verdict: default to HTTP APIs. They carry every feature a typical serverless or proxy backend needs at about a third of the cost. Reach for a REST API only when a specific feature in the right column is genuinely required, and then ideally on the one route that needs it rather than the whole API.

When should I use a REST API instead of an HTTP API?

Use a REST API only when you depend on a feature HTTP APIs do not provide. The REST-only list is short but real: built-in response caching, usage plans with API keys for metering partners, request and response validation and transformation, AWS WAF for managed rule protection, and private APIs exposed through an interface VPC endpoint. If your API needs none of these, the REST tier is pure overpayment. A common and costly mistake is standing up a REST API out of habit, using only Lambda proxy integration, and paying triple for years. For the protection layer specifically, weigh the cost in our guide on AWS WAF and Shield cost optimization before assuming you need REST for WAF.

The caching trap

REST API caching is billed per hour by cache size, not per hit, so an enabled cache with a poor hit ratio is paying continuously to save nothing. If you turn on caching, size it to a measured hit ratio, the same discipline that governs CloudFront cache hit ratio.

What else drives API Gateway cost besides requests?

Beyond the per-request charge, three things move the bill: data transfer, REST caching, and the backend. Data transfer out is billed at standard AWS rates and can exceed the request charge for APIs that return large payloads, so compress responses and avoid shipping fields the client discards. REST API caching adds a fixed hourly charge by cache size whether or not it is used. And the integration target, usually Lambda or a container, is billed separately, so an inefficient backend inflates the true cost of every call. For the serverless backend, our explainer on Savings Plans for Lambda and Fargate covers how to discount the compute behind the API.

Paying REST rates for an HTTP API workload?

Our AWS cost audit inventories every API Gateway stage, flags REST APIs that use no REST-only features, and models the saving from moving them to HTTP APIs, proven against a clean baseline. On the performance model, you pay only from realized savings. No savings, no fee.

Book a cloud cost audit →

How do I migrate a REST API to an HTTP API safely?

Migrate route by route, starting with the highest-traffic endpoints that use no REST-only features. Stand up the HTTP API alongside the existing REST API, point a weighted share of traffic at it through a custom domain or your load balancer, confirm parity on auth, CORS, and throttling, then shift the remainder. Keep any route that truly needs caching, API keys, WAF, or a private endpoint on REST, since a hybrid is fine and you only pay the REST rate on the routes that stay. The savings scale with traffic, so the highest-volume routes are where the migration pays for itself fastest.

Go deeper · free field guide

The AWS Cost Optimization Field Guide includes the API Gateway audit we run to separate routes that need REST from routes overpaying for it.

Frequently asked questions

How much cheaper are HTTP APIs than REST APIs?

HTTP APIs are roughly 71 percent cheaper than REST APIs. In US East, HTTP APIs are billed at about $1.00 per million requests for the first 300 million per month, while REST APIs are about $3.50 per million for the first 333 million. Verify current rates in your Region on the AWS pricing page.

When should I use a REST API instead of an HTTP API?

Use a REST API when you need features HTTP APIs do not offer: built-in response caching, usage plans and API keys, request and response validation, AWS WAF integration, private APIs over an interface VPC endpoint, or fine-grained request and response transformation. If you do not use those features, an HTTP API does the same job for about a third of the cost.

Does API Gateway have a free tier?

Yes. The AWS Free Tier includes one million REST or HTTP API calls per month for the first 12 months, plus one million WebSocket messages and 750,000 connection minutes. After 12 months, or above those volumes, standard per-million pricing applies.

What else drives API Gateway cost besides requests?

Data transfer out, optional REST API caching billed per hour by cache size, and any backing services such as Lambda or the integration target. REST API caching in particular is a fixed hourly charge whether or not the cache is hit, so size it to a real hit ratio or turn it off.

The short version

Amazon API Gateway HTTP APIs cost about $1.00 per million requests against $3.50 for REST APIs, roughly 71 percent cheaper, and carry every feature a typical serverless or proxy backend needs. Choose REST only for caching, usage plans, request validation, WAF, or private VPC endpoints, ideally on the single route that needs it. Watch data transfer and the per-hour REST cache charge, which bill regardless of hit ratio. When you want every API Gateway stage audited and the overpriced REST routes moved to HTTP, that is part of what our AWS cost optimization service delivers.

Primary sources & further reading

Cloud pricing and service behavior change frequently. Verify the specifics in this guide against the providers’ own current documentation and the FinOps Foundation: AWS pricing ↗, AWS documentation ↗ and FinOps Foundation Framework ↗. This article also reflects Cloud Cost Room’s hands-on, vendor-neutral engagement experience.

Written by Morten Andersen

Co-founder of Cloud Cost Room and a FinOps Certified Practitioner, with 20 years in IT and cloud cost optimization across AWS, Azure, Google Cloud and OCI. More about Morten →

More from the AWS Cost Optimization cluster

See every guide in the AWS Cost Optimization cluster →

The Cloud Cost Brief

Cloud pricing moves. We tell you when it matters.

New commitment instruments, FOCUS changes, hyperscaler pricing shifts, and the plays that actually move a bill. No schedule, no filler.

Subscribe · Work email only