# Fly Machines vs AWS Lambda MicroVMs: Firecracker by the machine or the baseline

Both run Firecracker. A Fly Machine is priced by size while started, a MicroVM by its baseline each second: $1.44 against $4.20 for the benchmark.

**Runtime costs 55% less than a Fly Machine and 85% less than a Lambda
MicroVM on it.** One thousand one-minute runs of a 2 vCPU, 4 GiB sandbox,
each with 20 CPU-seconds of work, are $1.44 on Fly Machines, $4.20 on AWS
Lambda MicroVMs and $0.64 on Runtime.

## At a glance

|           | Fly Machines                                       | AWS Lambda MicroVMs                          | Runtime                                               |
| --------- | -------------------------------------------------- | -------------------------------------------- | ----------------------------------------------------- |
| Isolation | Firecracker microVM                                | Firecracker microVM                          | Firecracker microVM, own kernel                       |
| Price     | `performance-2x` with 4 GB, $0.0861 an hour, `iad` | $0.2522 an hour for 2 vCPUs and 4 GB         | $0.03125 an hour waiting, $0.08 busy, for 2 and 4 GiB |
| Largest   | 16 performance vCPUs and 128 GB                    | A 4 vCPU, 8 GB baseline                      | 16 vCPUs and 64 GiB, paid                             |
| Idle      | Root disk is temporary; suspend at 2 GB or less    | Suspend keeps memory and disk, up to 8 hours | Pause keeps memory and processes, 1 to 365 days       |
| Regions   | 18                                                 | AWS regions                                  | One US region                                         |

## Same hypervisor, a third of the price

A performance vCPU on Fly is a whole core held for the Machine, and still the
shape costs about a third of Lambda's per hour. Lambda's premium buys AWS
integration and suspend with memory for up to eight hours; Fly's Machines
suspend only at 2 GB of memory or less.

## Cost for the same job

2 vCPU, 4 GiB, 60 seconds a run, 20 CPU-seconds of work, 1,000 runs:

```
Machines  Size   1,000 × 60 s / 3,600 × $0.0861                = $1.44

Lambda    CPU    1,000 × 60 s × 2 × $0.0000276944              = $3.32
          Memory 1,000 × 60 s × 4 × $0.0000036667              = $0.88
          Total                                                  $4.20

Runtime   CPU    1,000 × 20 s / 3,600 × $0.025                = $0.14
          Memory 1,000 × 60 s / 3,600 × 4 × $0.0075           = $0.50
          Total                                                  $0.64
```

Both rivals charge by size, so keeping both CPUs busy for the whole minute changes neither: $1.44 on Machines and $4.20 on Lambda, with $1.33 on Runtime.

## Which one to pick

- **Fly Machines** for long-running apps in many regions, alongside Fly's
  volumes and Postgres.
- **Lambda MicroVMs** for sandboxes tied into AWS identity and networking.
- **Runtime** for agent sandboxes at the lowest cost, with memory kept across
  pauses for up to a year.

See [Runtime vs Fly.io](/docs/fly-alternative) and
[Runtime vs AWS Lambda MicroVMs](/docs/aws-lambda-microvms-alternative).

## Sources

Fly Machines rates checked 23 September 2026; Lambda MicroVMs rates checked 25 September 2026.

- [Fly.io pricing](https://fly.io/pricing/)
- [AWS Lambda pricing](https://aws.amazon.com/lambda/pricing/), [Lambda MicroVMs](https://aws.amazon.com/lambda/lambda-microvms/)
- Runtime [pricing](/docs/pricing)

Facts on this page were checked on 25 September 2026.