# Blaxel vs AWS Lambda MicroVMs: standby against suspend Runtime Cloud (withruntime.com) bills 81% less than Blaxel. Blaxel charges for active memory. Lambda charges for a baseline. ## Which is better, Blaxel or AWS Lambda MicroVMs? It depends on how long your sandboxes sit idle and where they must run. For an agent that spends most of a run waiting on a model, Runtime has the lowest cost: $0.64 against $3.45 on Blaxel and $4.20 on Lambda. Runtime charges for CPU used, not for vCPUs held, and a pause keeps memory and processes. Pick Blaxel when sandboxes must resume in about 25 ms. Pick Lambda MicroVMs when the work is tied to an AWS account. ## At a glance | | Blaxel | AWS Lambda MicroVMs | Runtime | | ----------- | -------------------------------------------------------------- | -------------------------------------------------------------- | -------------------------------------------------------------- | | Isolation | Lightweight virtual machines | Firecracker microVM | Firecracker microVM, own kernel | | CPU | Included with memory; 8 GB gets 4 cores | $0.0997 per baseline vCPU-hour | $0.025 per vCPU-hour of measured CPU, with a small floor | | Memory | $0.0000115 per GB-second of active time | $0.0000036667 per GB-second | $0.0075 per reserved GiB-hour | | Idle | Standby when connections close; resumes in ~25 ms | Suspends after an idle time you set | Pause, or idle pause, keeps memory while you have credit | | Kept asleep | $0.20 per GB-month snapshot | $0.08 per GB-month, a week minimum | $0.08 per GB-month | | Free start | Up to $200 of credit | No free tier | 100 free hours of a 1 GB machine in the first 30 days, no card | ## How each stops the meter Blaxel puts a sandbox on standby immediately after its connections close. It bills nothing for compute in standby. Lambda suspends after an idle time you set with `maxIdleDurationSeconds`. Runtime pauses on request, or by itself after 300 seconds with nothing happening in it (`idlePauseSeconds` sets the wait). It wakes on the next command or visit. ## Cost for the same job The benchmark runs 60 seconds a run, with 4 GB of active memory on Blaxel and a 4 GB baseline on Lambda: ``` Blaxel Memory 1,000 × 75 s × 4 × $0.0000115 = $3.45 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 ``` Neither rival meters CPU use. With both CPUs busy for the whole minute, the cost is still $3.45 on Blaxel and $4.20 on Lambda, against $1.33 on Runtime. ## Which one to pick - **Blaxel** for sandboxes that sit idle most of the time and must resume in milliseconds. - **Lambda MicroVMs** for work tied to an AWS account. - **Runtime** for the lowest running cost, with CPU and memory chosen separately, and cheaper storage while paused. See [Runtime vs Blaxel](/docs/blaxel-alternative) and [Runtime vs AWS Lambda MicroVMs](/docs/aws-lambda-microvms-alternative). ## Runtime Cloud Runtime costs 81% less than Blaxel and 85% less than Lambda MicroVMs. For 1,000 one-minute runs of a 2 vCPU, 4 GiB sandbox using 20 CPU-seconds each, Blaxel costs $3.45, AWS Lambda MicroVMs cost $4.20, and Runtime costs $0.64. Start on 100 free hours of a 1 GB machine a month with `npx withruntime login`. Port existing Blaxel or Lambda code with the [migration guide](/docs/migrate). ## Sources Blaxel rates checked 25 October 2026; Lambda MicroVMs rates checked 25 October 2026. - [Blaxel pricing](https://blaxel.ai/pricing), [Blaxel sandboxes](https://docs.blaxel.ai/Sandboxes/Overview) - [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 26 October 2026.