AWS API Gateway Pricing: Models, Free Tier, and Cost Examples (2026)

Pricing overview

AWS API Gateway's pricing structure is designed around a pay-as-you-go model, meaning costs are incurred based on actual usage rather than fixed monthly fees, except for specific features. The primary components of AWS API Gateway costs are the number of API requests processed and the amount of data transferred out from the API Gateway. This model applies across its different API types: HTTP APIs, REST APIs, and WebSocket APIs, each with slightly varied pricing details reflecting their distinct functionalities and underlying infrastructure AWS API Gateway pricing page.

Additional costs can arise from optional features or integrated AWS services. For instance, caching for REST APIs incurs charges based on cache size. Custom domain names for REST APIs leverage AWS Certificate Manager, which is generally free for certificates provisioned for use with other AWS services, but the underlying infrastructure may have costs AWS Certificate Manager pricing. Private API endpoints, which allow API access only from within a Virtual Private Cloud (VPC), require a VPC endpoint, which has its own hourly charges and data processing fees AWS VPC pricing details. Moreover, integration with other AWS services like Lambda, DynamoDB, or S3 will incur their respective costs, as API Gateway acts primarily as a front door to these backend services.

Plans and tiers

AWS API Gateway does not operate on a traditional multi-tier plan system like 'Basic,' 'Pro,' or 'Enterprise.' Instead, its pricing is usage-based, with costs decreasing per million requests as volume increases. This constitutes a tiered pricing structure where the per-request cost is lower for higher volumes of API calls within a given month. This approach is consistent across HTTP, REST, and WebSocket APIs, though the specific rates differ. There are no upfront fees or termination charges.

HTTP API Pricing

HTTP APIs are designed for low-latency, cost-effective, and minimal-feature API proxies. They are generally the most affordable option for simple request-response scenarios.

• Requests: Priced per million requests, with volume discounts applied.
• Data Transfer Out: Standard AWS data transfer rates apply after a free tier.

REST API Pricing

REST APIs offer more features, including API keys, usage plans, request validation, and caching. They often integrate with AWS CloudFront for edge optimization by default.

• Requests: Priced per million requests, with volume discounts. This includes charges for API calls and any associated caching.
• Caching: Additional hourly charges for provisioned cache capacity.
• Data Transfer Out: Standard AWS data transfer rates apply.

WebSocket API Pricing

WebSocket APIs are used for real-time, bi-directional communication applications.

• Connection Minutes: Priced per million connection minutes.
• Messages: Priced per million messages sent and received.
• Data Transfer Out: Standard AWS data transfer rates apply.

The following table illustrates a simplified overview of the pricing components across different API Gateway types. Specific rates vary by AWS Region and are subject to change; refer to the official AWS API Gateway pricing page for current figures.

Free tier and limits

AWS offers a comprehensive free tier for API Gateway, primarily aimed at allowing new users to experiment and deploy small-scale applications without incurring costs during their first 12 months. This free tier is substantial and covers a significant volume of API calls and connection minutes across all API types. It's designed to help developers get started and test their applications before scaling AWS Free Tier details.

Free Tier Inclusions (for first 12 months):

• HTTP APIs: Up to 1 million API calls per month.
• WebSocket APIs: Up to 1 million API calls and 750,000 connection minutes per month.
• REST APIs: Up to 1 million API calls per month.
• Data Transfer Out: Up to 1 million messages for Data Transfer Out for REST APIs per month.

After the initial 12-month period, or once the free tier limits are exceeded within any given month, standard pay-as-you-go rates apply. It's important to monitor usage through the AWS Management Console to avoid unexpected charges, especially when transitioning beyond the free tier or scaling applications. The free tier is applied per account, not per API, meaning the total usage across all API Gateway instances within an account counts towards the monthly free tier limits.

Real-world cost examples

Understanding real-world costs for AWS API Gateway requires considering the API type, request volume, data transfer, and any additional features. These examples are illustrative and use typical pricing for the us-east-1 (N. Virginia) region as of early 2026. Always consult the official AWS pricing page for the most current and region-specific rates.

Example 1: Small HTTP API for a Mobile Backend

• Scenario: A simple mobile application backend using an HTTP API, receiving 5 million API calls per month. Minimal data transfer (e.g., 20 GB out).
• Calculation (approximate for us-east-1):
  • First 300 million requests: $1.00 per million.
  • 5 million requests * $1.00/million = $5.00
  • Data Transfer Out: 20 GB (assuming most is covered by a general AWS free tier for data out, or negligible cost for small volumes).
• Estimated Monthly Cost: Approximately $5.00 (plus potential minimal data transfer costs).

Example 2: Medium-Scale REST API with Caching

• Scenario: A REST API serving a web application, handling 50 million API calls per month. Caching enabled with a 0.5 GB cache (c0.5x instance type) for 24/7 operation. 100 GB of data transfer out.
• Calculation (approximate for us-east-1):
  • First 300 million requests: $3.50 per million.
  • 50 million requests * $3.50/million = $175.00
  • Caching (c0.5x): $0.02 per hour * 24 hours/day * 30 days/month = $14.40
  • Data Transfer Out: First 1 GB free. Next 9.999 TB at $0.09/GB.
  • 99 GB * $0.09/GB = $8.91
• Estimated Monthly Cost: Approximately $175.00 + $14.40 + $8.91 = $198.31.

Example 3: WebSocket API for a Real-time Dashboard

• Scenario: A real-time dashboard using a WebSocket API, maintaining 10,000 concurrent connections for 8 hours a day, 22 business days a month. Each connection sends/receives an average of 10 messages per minute. Negligible data transfer.
• Calculation (approximate for us-east-1):
  • Total Connection Minutes: 10,000 connections * 8 hours/day * 60 min/hour * 22 days/month = 105,600,000 connection minutes.
  • Connection Minutes Cost: $0.25 per million connection minutes.
  • 105.6 million connection minutes * $0.25/million = $26.40
  • Total Messages: 10,000 connections * 10 messages/min * 60 min/hour * 8 hours/day * 22 days/month = 1,056,000,000 messages.
  • Messages Cost: $1.00 per million messages.
  • 1056 million messages * $1.00/million = $1,056.00
• Estimated Monthly Cost: Approximately $26.40 + $1,056.00 = $1,082.40.

How the pricing compares

When evaluating AWS API Gateway's pricing against alternatives like Azure API Management, Google Cloud Apigee, or Kong Gateway, several factors come into play. Cloud-native API management solutions generally follow a usage-based or capacity-based pricing model, while self-hosted or open-source solutions like Kong Gateway often involve infrastructure costs and support subscriptions.

• Azure API Management: Azure's offering typically features tiered pricing plans (Developer, Basic, Standard, Premium) with fixed monthly costs for certain capacities, plus additional charges for gateway units, API calls, and data transfer. The entry-level tiers can be more predictable for smaller, consistent workloads, but high-volume usage might necessitate a comparison of per-call rates Azure API Management pricing.
• Google Cloud Apigee: Apigee is positioned as an enterprise-grade API management platform, offering a free evaluation tier and then moving into more substantial subscription-based models (Standard, Enterprise, Enterprise Plus) that bundle features and capacity. Apigee's pricing can be significantly higher than AWS API Gateway for equivalent request volumes due to its comprehensive feature set and focus on large-scale enterprise deployments Google Apigee pricing overview.
• Kong Gateway: Kong offers both an open-source self-managed version and a commercial enterprise version (Kong Konnect). The open-source version itself is free, but users bear the full cost of underlying infrastructure (VMs, containers, load balancers) and operational overhead. Kong Konnect, the SaaS offering, has pricing based on API calls and features, similar to cloud providers but often with different bundling strategies Kong Gateway pricing. For example, a key difference is that with self-managed Kong, you pay for the compute resources it runs on, whereas with AWS API Gateway, the compute is abstracted and included in the request price.

AWS API Gateway's granular, pay-per-request and pay-per-connection-minute model, combined with its substantial free tier, generally makes it a cost-effective choice for applications with fluctuating traffic patterns or those just starting out. For very high, consistent volumes, the tiered discounts can also make it competitive. However, for organizations requiring extensive API analytics, monetization features, or advanced developer portals out-of-the-box, the bundled offerings of Apigee or Azure API Management might present a more integrated, albeit potentially more expensive, solution. The choice often depends on the specific feature requirements, expected traffic, and the operational model (serverless vs. managed infrastructure).