Pick a type and the hours you really run it. See every purchase option, the break-even, the newer generation and the cheapest region.
Loading the catalog…
Price List API, Savings Plans offer files, spot history and EC2 specs. Nothing typed in by hand.
Consistency rules run on every region file before it replaces the previous one. Catalog 2026-09-22.
No tax, EBS, data transfer or private pricing. Full methodology.
TruCost reads your actual AWS usage and tells you which of these commitments you should really buy, what your idle capacity costs, and where every rupee of the bill goes. Thirty minutes on your own data.
An AWS instance has four prices, not one. On-demand is the list price per hour, billed per second, with no commitment. Reserved instances are a one- or three-year commitment to a type in a region (EC2 can be convertible; RDS, ElastiCache, OpenSearch and Redshift reserve a specific class), paid all up front, partly up front or monthly, at 30 to 70 percent below on-demand. Savings Plans commit to an hourly spend instead of a type: Compute Savings Plans follow you across families, sizes, regions and even Fargate and Lambda; EC2 Instance Savings Plans are cheaper but tied to a family in a region; Database Savings Plans cover RDS, Aurora, ElastiCache and more from generation 7 onwards. Spot is spare capacity at up to 90 percent off, with two minutes' notice before AWS takes it back.
The right answer depends on hours. A commitment is paid whether the instance runs or not, so an instance that runs business hours only, roughly 200 hours a month, rarely earns back a reservation, while one that runs all month almost always does. That is why every price on this site is shown at the hours you choose, with the break-even marked, rather than as a per-hour list. For the reasoning behind each purchase option, read our guide to the AWS Pricing Calculator and the complete guide to Savings Plans and reserved instances.
Yes. No login, no email, no limits. Prices are refreshed nightly from the AWS Price List API and the Savings Plans offer files, and every page states when it was last updated.
AWS's calculator estimates a whole architecture: you add each service and its usage and get a monthly total. This one answers the question engineers ask most: what does one instance cost at the hours I really run it, and which commitment pays off. It compares on-demand, 1- and 3-year reserved instances, Compute and EC2 Instance Savings Plans and spot side by side, shows the break-even hours, the newer generation that costs less and the cheapest region for the same type.
EC2, RDS (every engine, single-AZ and multi-AZ), ElastiCache, OpenSearch and Redshift, in every commercial AWS region. Detail pages exist for every type in the primary region, and the table pages cover every region.
The number of hours a month from which a reserved instance or Savings Plan costs less than paying on-demand for the same instance. Below that many hours, on-demand is cheaper; above it, the commitment wins. Run the estimate at your real hours rather than the 730-hour default.
Spot prices are the recent average from the Spot price history for the region and type, with a note on how much they moved. Spot capacity can be interrupted, so treat spot as a discount on interruptible work, not a replacement for a reserved baseline.
Yes. Add lines to the estimate with different purchase options, see the monthly and annual total in USD, INR or EUR, and share the estimate as a link.
Every nightly build re-fetches a sample of prices from AWS and fails if anything drifts, checks invariants such as reserved never exceeding on-demand, and cross-checks EC2 against an independent public dataset. The methodology page describes each step.
Tell us a little about your AWS setup and we will come back within one business day — or pick a slot straight away.