∑
DevFormula
AI Tokens & Cloud ComputeVerified 2026-10-21

AWS Lambda ARM Graviton2 Cost & Savings Calculator

Simulate AWS Lambda ARM64 compute bills ($0.20/1M requests + $0.0000133334 per GB-second) with 20% lower unit rates.

Parameters & Config

calls
calls
1,000calls1,000,000,000calls

Total function executions triggered per month

MB
MB
128MB10,240MB

RAM allocated to the serverless container

ms
ms
1ms900,000ms

Average latency per execution in milliseconds

$
$
0.01$5$

Cloud provider fee per 1,000,000 invocation requests

$
$
0$0.001$

Charge per GB-second of compute runtime consumed

Total Monthly Cloud BillLIVE ESTIMATE
$8.67

Combined monthly charge for invocations and compute runtime

Effective Cost per 1M CallsLIVE ESTIMATE
$0.87

Blended cost to execute 1,000,000 function requests

Key Metrics & Sub-Totals

3 items
Compute Runtime ChargeCost based on memory allocated and total execution duration (GB-seconds)
$6.67
Invocation Request ChargeCost for incoming API function invocations
$2.00
Monthly GB-Seconds ConsumedTotal memory-time compute units utilized
500,000 GB-s

Calculation Summary & Unit Breakdown

MetricCalculated ValueNotes / Logic
Monthly Invocations10,000,000requests/month
Allocated Function Memory512 MB0.500 GB
Average Execution Duration100 msper call
Total Compute Volume500,000 GB-sGB-seconds
Request Surcharge$2.00Invocation fee
Compute Runtime Charge$6.67Memory-duration fee
Effective Cost per 1M Calls$0.87Blended rate
Total Monthly Cloud Bill$8.67USD

Overview & Methodology

AWS Graviton2 processors (ARM64 architecture) deliver up to 19% better performance at 20% lower cost compared to standard x86 processors: $0.20 per 1M requests and $0.0000133334 per GB-second.

Mathematical Formulation

Compute Cost = Total GB-Seconds × $0.0000133334. Total Monthly = Invocation Cost + Compute Cost.

Optimization Best Practices

1

Switching architecture in SAM, Serverless Framework, or Terraform requires just setting `Architectures: [arm64]`.

2

Python, Node.js, and Java runtimes run on ARM64 with zero code modifications.

3

Graviton2 features higher memory bandwidth, often resulting in shorter runtime durations.

Frequently Asked Questions

Q.How much does 10M executions cost on Graviton2 at 512MB and 100ms?
A.

10M requests cost $2.00 invocation + $6.67 compute (500,000 GB-s), totaling $8.67/month (saving $1.66 vs x86).

Q.Do all AWS Lambda runtimes support ARM64?
A.

Yes. Python, Node.js, Java, Go, Rust, .NET, and custom Docker container images all support ARM64.

Q.Does the AWS Lambda Free Tier apply to Graviton2?
A.

Yes. The 400,000 GB-seconds monthly free tier applies equally to Graviton2 functions.