Billing Commitment
Lambda@Edge is billed by request and execution duration; CloudFront distribution, request, and data-transfer charges remain separate.
AWS Lambda@Edge Pricing, Features & Alternatives
Node.js and Python functions to customize CDN requests and responses across globally distributed edge locations.
AWS Lambda@Edge runs Node.js and Python functions in response to CloudFront viewer or origin request and response events. Functions are created in the US East (N. Virginia) region and replicated to AWS locations closer to users; the applicable execution limits and supported features depend on the trigger. Unlike CloudFront Functions, Lambda@Edge can handle origin-facing logic and more substantial request processing.
Best for
Request and response customization that needs more runtime capability, network access, or execution time than CloudFront Functions provides.
Watch out for
Functions must be created in us-east-1, replicated globally by CloudFront, and comply with event-specific limits and unsupported Lambda features.
- Workloads
- Edge Request Processing
- Services
- CDN
- Edge Functions
- Delivery
- Managed
Development & Execution
Supported platform capabilities, execution environments, development technologies, deployment models, integrations, and triggers.
Runtimes
- Node.js
Languages
- JavaScript
- Python
- TypeScript
Capabilities
- Compute Autoscaling
Architecture
- x64
Deployment & packaging
- Serverless (Managed Runtime)
Integrations
- Terraform
Triggers
- Edge Events
- HTTP Requests
Pricing
Published pricing structure, representative costs, and material billing factors.
- Starting Price
- $0.60
- Primary Billing
- Resource based
- Commitment
- Lambda@Edge is billed by request and execution duration
AWS Lambda@Edge Usage
Primary PricingLambda@Edge function execution charges; CloudFront distribution and delivery charges are separate.
| Resource | Billing Unit | Price |
|---|---|---|
| Lambda@Edge requests | per 1 million requests | $0.60 |
| Lambda@Edge duration | per 128 MB-second | $0.00000625125 |
Infrastructure & Availability
Where workloads run, how deployment is scoped, and what underlying infrastructure the platform depends on.
Lambda@Edge functions are authored and maintained in the AWS US East (N. Virginia) Region, us-east-1. After association with a CloudFront distribution, AWS replicates and runs them at AWS locations closer to viewers. AWS does not publish a current product-specific execution-location count.
Execution
- Edge
Deployment scope
- Global
Infrastructure providers
- Amazon Web Services
Availability reviewed · Availability source
Resource & Execution Limits
Memory
128 MB / 10,240 MB
Viewer events / origin events
Execution time
30 seconds
Per invocation
Function package
50 MB compressed
Per function
Request rate
10,000 requests/second
Per Region
Security & Compliance
Security Controls
- Access Control
- IAM permissions, execution roles, and service-linked roles Lambda@Edge
- Audit Logging
- CloudTrail management events and CloudWatch edge logs Edge logs are best effort
FAQ
Does Lambda@Edge run on every CloudFront request?
Not necessarily. Viewer-request and viewer-response functions run for each applicable viewer request or response. Origin-request and origin-response functions run only when CloudFront forwards the request to the origin, such as after a cache miss; if CloudFront serves a valid cached object, those origin-event functions do not run.
Does Lambda@Edge support all AWS Lambda features?
No. Lambda@Edge does not support several standard Lambda features, including VPC access, custom environment variables, Lambda layers, X-Ray tracing, provisioned concurrency, container-image functions, or arm64 functions.
Is Lambda@Edge free?
No permanent free Lambda@Edge compute allowance is included with CloudFront pricing plans. Lambda@Edge invocations are billed on a pay-as-you-go basis, separately from CloudFront plan charges.
How much memory can Lambda@Edge use?
Lambda@Edge memory depends on the event type. Viewer-request and viewer-response functions are limited to 128 MB, while origin-request and origin-response functions can use up to 10,240 MB (10 GB).
