Pricing
Flexible pricing for any scale
Pick a plan, pay for the compute you actually use. Scale-to-zero means idle costs nothing.
Free
$0/ month
Try the platform on real workloads — microVMs, not a sandbox.
- 3 apps · 1 region
- 512 MB · 1 vCPU per microVM
- 100k invocations / month
- Snapshot wake · shared pool
- 7-day logs
Pro
popular$29/ month
For production services that must wake fast and stay isolated.
- Up to 100 apps · 3 regions
- Up to 4 vCPU · 4 GB per microVM
- 5M invocations / month
- Priority wake pool · p50 under 350 ms
- Cron, queues with DLQ, delayed tasks
- 30-day logs, traces and wake timelines
Scale
$99/ month
Dedicated bare metal, custom egress policy and an audit trail.
- Dedicated hosts · all regions
- Up to 16 vCPU · 32 GB per microVM
- Committed-use invocation pricing
- Reserved warm capacity
- SSO, audit log, custom egress allowlists
- 99.95% SLA and support channel
Questions
Nothing. An idle app is a snapshot on disk — no vCPU, no memory, no billing. You pay for wake time and invocation duration only.
No. A cold start boots a runtime from scratch. A voopr wake restores the exact memory snapshot of your already-warm process, which is why the documented p50 is under 350 ms.
Yes, on Pro and Scale you can reserve warm capacity per app; requests then skip the restore step entirely.
Invocations above the plan allowance are billed per million at the plan's overage rate. Nothing gets throttled silently.