# Freestyle vs Prime Sandboxes: pricing, forks and concurrency Runtime Cloud (withruntime.com) costs 58% less than either. Freestyle forks live VMs. Prime runs a thousand at the same time. ## Which is better, Freestyle or Prime Sandboxes? It depends on how big each sandbox is and how many run at the same time. If your agent mostly waits on model replies, Runtime is the cheapest here: $0.64 for 1,000 one-minute runs, against $2.25 on Freestyle and $1.52 on Prime. It also has forks and pauses that keep memory, and no plan fee. Pick Freestyle for VMs of up to 32 vCPUs that fork without a pause. Pick Prime Sandboxes for rollouts by the thousand inside Prime Intellect. ## At a glance | | Freestyle | Prime Sandboxes | Runtime | | --------- | ------------------------------------------------------------------------ | ---------------------------------------------- | -------------------------------------------------------------- | | Isolation | Full Linux virtual machines | Hardware-virtualized microVM, own guest kernel | Firecracker microVM, own kernel | | CPU | $0.04032 per allocated vCPU-hour | $0.02 per vCPU-hour while running | $0.025 per vCPU-hour of measured CPU, with a small floor | | Memory | $0.0129 per GiB-hour | $0.0125 per GiB-hour | $0.0075 per reserved GiB-hour | | Forks | Live, without pausing the VM | From a filesystem checkpoint; no memory | Up to 10 copies of a running sandbox | | At once | 10 free, 40 Hobby, 400 Pro | 1,024 per account to start | 100 paid | | Plan | Free, Hobby $50, Pro $500 a month | None; rates through 22 December 2026 | None; prepaid credit from $10 | ## Forks today, or scale today Freestyle's strength is state: live forks and weeks-long hibernation. Prime's strength is breadth: a thousand microVMs per account from the start, at about half Freestyle's CPU rate, with forks that copy files only. Runtime has forks and memory-keeping pauses now, and prices waiting time at a floor of a twentieth of a vCPU. ## Cost for the same job The benchmark is 60 seconds a run, with each rival's default disk (32 GiB on Freestyle, 5 GiB on Prime): ``` Freestyle CPU 1,000 × 60 s / 3,600 × 2 × $0.04032 = $1.34 Memory 1,000 × 60 s / 3,600 × 4 × $0.0129 = $0.86 Disk 1,000 × 60 s / 3,600 × 32 × $0.000086 = $0.05 Total $2.25 Prime CPU 1,000 × 60 s / 3,600 × 2 × $0.02 = $0.67 Memory 1,000 × 60 s / 3,600 × 4 × $0.0125 = $0.83 Disk 1,000 × 60 s / 3,600 × 5 × $0.0002 = $0.02 Total $1.52 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 ``` At 100,000 runs a month, the cost is $224.99 on Freestyle, $151.67 on Prime, and $63.89 on Runtime. With both CPUs busy for the whole minute: $2.25 on Freestyle, $1.52 on Prime, and $1.33 on Runtime. ## Which one to pick - **Freestyle** for VMs of up to 32 vCPUs that fork without a pause. - **Prime Sandboxes** for rollouts by the thousand inside Prime Intellect. - **Runtime** for the lowest price per run, forks and pauses that keep memory, and no plan fee. See [Runtime vs Freestyle](/docs/freestyle-alternative) and [Runtime vs Prime Sandboxes](/docs/prime-sandboxes-alternative). ## Runtime Cloud Runtime costs 72% less than Freestyle and 58% less than Prime on this benchmark. The benchmark is 1,000 one-minute runs of a 2 vCPU, 4 GiB sandbox, with 20 CPU-seconds of work in each run: $2.25 on Freestyle, $1.52 on Prime Sandboxes and $0.64 on Runtime. Runtime bills the CPU your code uses, in a Firecracker microVM per sandbox. Sign in with `npx withruntime login` to start free. Use the [migration guide](/docs/migrate) to bring existing code. ## Sources Freestyle and Prime rates were checked 25 October 2026. - [Freestyle pricing](https://www.freestyle.sh/pricing), [Freestyle VM pricing and limits](https://www.freestyle.sh/docs/vms/pricing-and-limits) - [Prime Sandboxes overview](https://docs.primeintellect.ai/sandboxes/overview), [Prime Sandboxes launch](https://www.primeintellect.ai/blog/sandboxes) - Runtime [pricing](/docs/pricing) Facts on this page were checked on 26 October 2026.