AWS EC2 Pricing Calculator
Estimate AWS EC2 monthly and annual cost across instance types, regions, and pricing models. Covers t3, t4g, m5, m6i, m7i, c5, c6i, c7i, r5, r6i, and r7i families with On-Demand, 1-Year and 3-Year Reserved Instances, and Compute Savings Plans. Includes optional EBS gp3 storage and outbound data transfer for a realistic total bill estimate. Approximate pricing snapshot for 2026; verify final budget at calculator.aws before committing.
AWS EC2 monthly cost equals hourly rate times 730 hours times number of instances, plus EBS storage at $0.08 per GB and outbound transfer at $0.09 per GB above 100 GB free. For 3 m5.large instances on Linux in us-east-1 with 1-Year Reserved Instances, the monthly bill is about $134 in compute, or $1,615 per year, saving $908 versus On-Demand.
Enter Values
Family abbreviations: t = burstable, m = general, c = compute, r = memory. Graviton (t4g) is ARM-based and runs Linux workloads cheaper.
Windows adds a license premium of roughly 50 to 100% on top of Linux pricing.
730 = 24 hours x 365 days divided by 12. Use 220 for business hours only.
Optional: gp3 SSD volume size attached to your instances.
Optional: outbound to internet. First 100 GB per month is free.
Result
Enter values above and click Calculate to see your result.
AI Assistant
Ask about this calculator
I can help you understand the aws ec2 pricing calculator formula, interpret your results, and answer follow-up questions.
Try asking
Formula
How it works: Each EC2 instance has a base On-Demand hourly rate that varies by instance type and region. Reserved Instances and Savings Plans apply a discount factor to that rate (typically 28 to 52% off depending on term and commitment). Multiply the effective hourly rate by hours per month (730 for 24/7 use) and instance count to get monthly compute cost. Add EBS gp3 storage at about $0.08 per GB per month and outbound data transfer at $0.09 per GB above the 100 GB monthly free tier.
Worked Example
How AWS EC2 Pricing Actually Works
EC2 pricing has four main levers: instance type, region, operating system, and commitment level. The instance type sets a base hourly rate that scales with vCPU and memory. A t3.micro starts at about 1 cent per hour, an m5.large around 10 cents per hour, and an r5.4xlarge over $1 per hour. Each region applies its own multiplier on top, with US regions cheapest and Asia Pacific regions 20 to 30% more expensive.
The biggest cost lever for production workloads is the pricing model. On-Demand has no commitment but pays full rate. Reserved Instances commit you to a specific configuration for 1 or 3 years for a 36 to 52% discount. Compute Savings Plans commit to an hourly spend (any instance family, region, or OS) for 28 to 50% off. Spot Instances run on spare AWS capacity at up to 90% off but can be interrupted on 2 minutes notice.
For most production workloads at steady-state, a 1-Year Compute Savings Plan is the safest first commitment: 28% off On-Demand and full flexibility to change instance types or regions during the term. Once a workload pattern is fully understood after a year, upgrading to 3-Year Reserved or Savings Plan unlocks the deeper 50%+ discounts.
Do not forget the secondary costs. EBS gp3 storage at $0.08 per GB per month adds up fast on large fleets. Outbound data transfer to the internet is $0.09 per GB after the first 100 GB and is the most commonly underestimated EC2 cost. A high-traffic web app or video service can spend more on egress than on compute.
- On-Demand sets the baseline rate; every other model discounts off it.
- 1-Year Reserved Instances save about 36%, 3-Year Reserved save about 52%.
- Compute Savings Plans trade some discount for flexibility across instance types.
- Spot Instances save up to 90% but can be reclaimed with 2 minutes warning.
- EBS gp3 storage is $0.08 per GB per month; outbound transfer is $0.09 per GB after 100 GB free.
- Asia Pacific regions cost 20 to 30% more than US regions for the same workload.
- Graviton instances (t4g, m6g, c6g, r6g) save about 20% on Linux workloads.
For a quick ballpark, this calculator covers 90%+ of common EC2 configurations. For Spot Instance modeling, GPU instance types, Dedicated Hosts, or final budget commitments, use the official AWS Pricing Calculator at calculator.aws which pulls live prices from the AWS API. For broader cloud cost analysis on AI workloads, see the AI Chatbot Cost Calculator and Agentic ROI Calculator.
You can also calculate changes using our AI Chatbot Cost Calculator, Agentic ROI Calculator, SaaS Runway Calculator, Workforce Cost Calculator or Cost Per Hire Calculator.
EC2 Pricing Model Comparison (Approximate Discount vs On-Demand)
Typical discount levels for the most common EC2 commitment tiers. Use these to plan workload commitments against expected steady-state usage.
| Pricing Model | Term | Discount | Best For |
|---|---|---|---|
| On-Demand | None | 0% | Spiky workloads, dev and test, short-lived jobs |
| 1-Year Reserved (No Upfront) | 12 months | ~36% | Steady predictable load, locked instance family |
| 3-Year Reserved (No Upfront) | 36 months | ~52% | Long-term stable workloads, max RI savings |
| 1-Year Compute Savings Plan | 12 months | ~28% | Mixed instance flexibility, region or OS changes likely |
| 3-Year Compute Savings Plan | 36 months | ~50% | Long-term flexibility across families and regions |
| Spot Instances | None | up to 90% | Fault-tolerant batch, CI workers (not modeled here) |
Note: Approximate values; actual discount varies by instance family and region. All Upfront and Partial Upfront commitments save another 2 to 10 percentage points. Verify exact numbers at calculator.aws before committing.
Frequently Asked Questions
How is AWS EC2 pricing calculated?
EC2 pricing combines four variables: instance type, region, operating system, and pricing model. The base On-Demand rate is set per instance family per region. Multiply that hourly rate by 730 hours for a 24/7 month, then by the number of instances. Reserved Instances and Savings Plans apply a discount factor of 28 to 52% on the On-Demand rate depending on term length and commitment type. Add EBS storage and outbound data transfer for the full bill.
What is the difference between Reserved Instances and Savings Plans?
Reserved Instances commit to a specific instance family in a specific region for 1 or 3 years and lock in a discount of 36 to 52%. Compute Savings Plans commit to an hourly compute spend for 1 or 3 years (any region, family, OS, or tenancy) at 28 to 50% off. Reserved Instances offer slightly deeper discounts on a fixed config; Savings Plans trade some discount for flexibility. Both bill at the discounted rate automatically when matching usage occurs.
How much can I save with Reserved Instances?
Typical savings vs On-Demand: 1-Year No Upfront Reserved Instance saves about 36%. 3-Year No Upfront saves about 52%. 1-Year All Upfront saves about 38%. 3-Year All Upfront saves up to 63% but requires the full payment at start. For an m5.large running 24/7 in us-east-1, that is $25 to $44 per instance per month in savings. The break-even on a 1-year RI is roughly 8 to 9 months of usage.
How much does data transfer cost on EC2?
Inbound data transfer to EC2 is free. Outbound to the internet has a 100 GB per month free tier across all AWS services, then about $0.09 per GB up to 10 TB per month, dropping to $0.085, $0.07, and $0.05 per GB at higher volume tiers. Asia Pacific regions like Singapore and Tokyo charge roughly 33% more for outbound transfer. Inter-region transfer is billed at the source region rate.
Why is Windows on EC2 more expensive than Linux?
Windows includes a Microsoft license fee that AWS pays Microsoft and passes through. The premium is roughly $0.046 per vCPU per hour for general and compute instances and lower for burstable T-family. For an m5.large with 2 vCPUs, that adds $0.096 per hour over Linux, doubling the price. SQL Server licenses add even more on top. Linux (especially Amazon Linux 2023) is the cheapest path on EC2.
Should I use Graviton (t4g, m6g) for cost savings?
Graviton is AWS-built ARM-based silicon. It runs Linux workloads at roughly 20% lower cost per hour and often delivers better price-performance for web servers, microservices, containers, and most general workloads. Graviton requires ARM-compatible builds; most modern languages (Go, Node, Python, Java) handle this automatically. Graviton does not run Windows. If you can recompile or use ARM container images, Graviton is the cheapest path on EC2.
How accurate is this calculator vs the official AWS Pricing Calculator?
This calculator uses an approximate price snapshot covering common configurations and is meant for fast ballpark estimates. It is typically within 5% of the official numbers for the instance types and regions included. For final budget commits, especially for Spot Instances, Dedicated Hosts, GPU instances, or specialized regions, use the official AWS Pricing Calculator at calculator.aws which pulls live pricing from the AWS API.
Is it possible to embed the AWS EC2 Pricing Calculator on another website?
Yes, embedding the AWS EC2 Pricing Calculator is free. Hit the "Embed" button on this page, adjust the width, height, and theme, then grab the iframe code. It works on WordPress, Wix, Squarespace, Shopify, and plain HTML pages. No registration needed. Full instructions at calculory.com/services/embed-calculators.
AI Assistant
Ask about this calculator
I can help you understand the aws ec2 pricing calculator formula, interpret your results, and answer follow-up questions.
Try asking
More Business Calculators
View allProfit Maximizing Output Calculator
Find the output level that maximizes profit.
Website Value Calculator
Professional valuation for websites and online businesses.
Web Design Cost Calculator
Estimate website design costs with traditional and AI development options.
App Development Cost Calculator
Estimate mobile app development costs by scope and features.
Related Articles
All articles
Agentic ROI Blueprint: Replacing Roles in 2026
In 2026, teams are measuring total capability, not just headcount. Learn how to model the cost of replacing or augmenting full-time roles with an AI agent stack.
Read article
GST for AI Freelancers: Foreign Inward Remittance Guide 2026
AI freelancers serving global clients in 2026 can treat eligible work as zero-rated export of services under GST. Learn LUT, IGST refund, RCM on AI tools, and compliance timelines.
Read article
Circular TCO vs Circular Savings: Payback Blueprint 2026
Compare linear ownership costs against circular savings with a practical 2026 model. Calculate annualized cost, lifetime savings, and payback months for sustainable investments.
Read article
COLA in 2026: How to Negotiate a Cost of Living Raise
A 3% raise with 4% inflation is a pay cut. Learn how to calculate your personal inflation rate and negotiate a COLA raise with data-driven leverage in 2026.
Read articleModern Tools for Every Need
Secure and Private
All calculations run locally. Your business data never leaves your browser.
Verified Precision
Precise Business Calculations Powered by Calculory AI