Vercel Sandbox vs Freestyle

Vercel Sandbox runs short jobs in Firecracker microVMs across 19 regions, and each session stops after 24 hours. Freestyle gives each agent a full KVM VM with systemd and nested virtualization that can run with no time cap, pause with its memory and wake on the next request.

The companies and products

Vercel Sandbox

Vercel is a cloud platform for web apps and agents, and Vercel Sandbox is its product for running untrusted or agent-written code in isolated Linux microVMs.

Each sandbox is a Firecracker microVM with its own kernel, started from a Vercel image, your own OCI image, or a saved snapshot. You drive it with the JavaScript or Python SDK or the sandbox CLI: run commands, move files, open ports, and set network rules. It is aimed at code interpreters, app builders, test runs and agent tools, and Vercel's docs say it is not meant to run nonstop.

Freestyle

Freestyle sells full Linux virtual machines for AI agents, built for long running, complex tasks.

Each Freestyle VM is a KVM virtual machine with its own kernel, booted from an Ubuntu base snapshot with systemd, Docker and coding agent CLIs preinstalled. VMs can be paused with their memory, snapshotted while running, forked and woken by the next request, and every VM starts behind a default-deny firewall. It is built for teams that give each agent or each user its own machine, driven from a TypeScript SDK, a CLI or an API.

Which use cases each one fits

For each use case, Battleships prices every sandbox provider on the same workload and sorts them from the cheapest monthly bill up. “22nd cheapest of 345” means 21 of the other 344 providers run that workload, with every feature it needs, for less. “With limits” means the provider runs it but misses something, so it is listed after every full match. Open a use case for its whole ranking and what each provider misses.

Use case and its workloadVercel Sandbox: price rankFreestyle: price rankbochka: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Neither has a desktop or browser for an agent to check the app it built, and Vercel also stops each session at 24 hours; bochka ranks 3rd with both built in.
244th of 367, with limits
$3,351 a month
241st of 367, with limits
$3,311 a month
3rd cheapest of 367
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
Freestyle VMs can run all month with no session cap, while Vercel's docs say sandboxes are not designed to run continuously; both are a compromise here, and bochka ranks 5th.
257th of 366, with limits
$3,657 a month
263rd of 366, with limits
$4,195 a month
5th cheapest of 366
$657 a month
Code interpreter bursts
100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month
Vercel Sandbox ranks 22nd and Freestyle ranks 40th: Vercel bills CPU only while code uses it, and Freestyle can pause a VM with its memory and wake it on the next request.
22nd cheapest of 345
$216 a month
40th cheapest of 345
$482 a month
186th of 345, with limits
$2,000 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
15th cheapest of 361
$1,267 a month
16th cheapest of 361
$1,358 a month
Cheapest of 361
$209 a month
RL rollouts
1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month
Vercel Sandbox ranks 3rd thanks to very high concurrency on Pro, while Freestyle Pro caps at 400 VMs at once, though it can fork a running VM with its memory.
3rd cheapest of 360
$15,761 a month
107th of 360, with limits
$6,600 a month
34th of 360, with limits
$2,000 a month
Agent evals
250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month
69th cheapest of 364
$1,508 a month
53rd cheapest of 364
$1,046 a month
10th cheapest of 364
$200 a month
CI runners
40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month
Vercel fits CI runs fully, while Freestyle's own docs send sustained heavy CPU work elsewhere; bochka ranks 4th and did about twice Vercel's CPU work per second.
107th cheapest of 377
$1,144 a month
168th of 377, with limits
$625 a month
4th cheapest of 377
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A dev box that stays up all day fits Freestyle, while on Vercel it restarts every 24 hours and loses its running processes.
276th of 368, with limits
$145 a month
98th cheapest of 368
$151 a month
18th cheapest of 368
$26 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
Neither ships a desktop, so on both you build the screen, VNC and input layer yourself; bochka ranks 6th with a 60fps desktop and computer use built in.
235th of 369, with limits
$952 a month
234th of 369, with limits
$940 a month
6th cheapest of 369
$200 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
Vercel has no browser feature and Freestyle has a guide to install Chromium and a display yourself; bochka ranks 7th with Chromium ready in every sandbox.
259th of 367, with limits
$486 a month
257th of 367, with limits
$482 a month
7th cheapest of 367
$200 a month
Scraping & bots (startup scale)
50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month
277th of 346, with limits
$468 a month
278th of 346, with limits
$482 a month
6th cheapest of 346
$90 a month
Mass scraping & bots (undetectable)
1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month
234th of 338, with limits
$9,960 a month
183rd of 338, with limits
$6,560 a month
71st of 338, with limits
$1,800 a month
Parallel red teaming
200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month
283rd of 360, with limits
$554 a month
272nd of 360, with limits
$482 a month
2nd cheapest of 360
$200 a month
GPU eval jobs
10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month
Can't run itCan't run itCan't run it
macOS agents
5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month
Can't run itCan't run itCan't run it

Key features

FeatureVercel SandboxFreestylebochka
IsolationFirecracker microVM with its own kernelFull KVM VM with its own kernel on bare metal serversFull VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
CPU benchmark (4 vCPU / 8 GB)12.96 runs/s9.64 runs/s26.69 runs/s
Longest session24 hours on Pro and Enterprise, 45 minutes on Hobby; resume starts a new sessionNo cap; runs until you pause, stop or delete it, or set your own limitNo cap on paid plans (--no-auto-stop); default TTL 1 hour, changeable
Pause and resumeFilesystem snapshot on stop; processes and memory start overPause saves memory and disk; the same processes resume where they left offDisk saved every minute and on stop; resume boots the machine again
Fork and wakeNew sandboxes boot from a saved filesystem snapshot; no wake on requestClone a running VM with its memory; a paused VM wakes on the next requestFork a running sandbox with its disk; a stopped sandbox is started by API
Docker and nested virtualizationDocker with sudo; no nested virtualizationDocker from boot, systemd as pid 1, full KVM supportDocker, root and systemd, with /dev/kvm for Android emulators and VMs
Desktop, browser, computer useNone; headless Linux onlyGuides to install GNOME or Xvfb, VNC and Chromium yourself60fps desktop and Chromium, computer-use and browser-control APIs, embeddable stream
Coding agentsClaude Code, Codex, OpenCode and pi preinstalled; you start them as commandsclaude, codex, opencode and openclaw preinstalled, driven through exec or a terminalSeven agents preinstalled, with an API to prompt, steer and stream events
Custom imagesYour own OCI images from a Dockerfile, via Vercel Container RegistryYour own snapshots built on the Ubuntu base; no OCI imagesNamed snapshots (templates), shareable by code; no Dockerfile builds
Biggest machine8 vCPU / 16 GB on Pro, 32 vCPU / 64 GB on Enterprise32 vCPU / 64 GiB on Pro, 64 / 128 with custom limitsUp to 16 vCPU / 32 GB on the $100 plan and above
Inbound accessPublic URLs for up to 15 ports, no built-in auth; no raw TCP or public IPHTTPS on free style.dev names or your own domain, forward auth, other public protocolsHTTPS URLs for up to 50 ports, token-protected by default; raw TCP; dedicated IPv6 or IPv4
Outbound rulesAllow-all, deny-all or a domain and CIDR allowlist, changeable while runningDefault-deny firewall with allow rules by IP, VM, VPC or exact domainOpen internet, no provider-side allowlist; firewall inside the VM
Keys kept outside the sandboxCredentials brokering injects headers for any HTTPS domain you allowCredentials injected at the edge for HTTPS headers, Postgres and S3Credentials proxy for model providers and any HTTPS API; switch a credential to another account when one runs out
Regions19 regions across the US, Europe, Asia, South America, Africa and AustraliaNot listed; the docs name a San Jose, US data centerEU only: Germany, Finland, France, with zero data retention
Every feature Battleships tracks (43)
FeatureVercel SandboxFreestylebochka
Full VM (own kernel)YesYesYes
Docker insideYesYesYes
Nested virtualizationNoYesYes
RootYesYesYes
SystemdNot publishedYesYes
Desktop GUINoNoYes
BrowserNoNoYes
Computer-use APINoNoYes
SSHNoYesYes
Persistent diskYesYesYes
Pause/resumeYesYesYes
Automatic snapshotsYesYesYes
Fork/cloneYesYesYes
Live fork (no pause)YesYesYes
Memory forkNoYesNo
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesYesYes
Idle auto-stopNoYesNo
Wake on requestNoYesNo
HTTPS preview URLsYesYesYes
Raw TCP inboundNoYesYes
Public IPv4NoNoNo
Egress allowlistYesYesNo
Secret proxyYesYesYes
Secret proxy for any APIYesYesYes
Preinstalled agentsYesYesYes
Agent harness APINoNoYes
MCP serverYesNoNo
GPUNoYesNo
WindowsNoNoNo
Arm64NoNoNo
Self-hostingNoNoNo
BYOCNoNoNo
Open sourceNoNoNo
Scoped API keysNot publishedYesYes
Spend limitsYesNot publishedYes
Usage per sandbox (API)NoNoYes
Audit logsNot publishedYesNo
SSOYesYesNo
SOC 2YesNot publishedNo
HIPAAYesNot publishedNo
EU data residencyYesNoYes
Zero data retentionNot publishedNot publishedYes

Performance

Vercel Sandbox12.96 runs/s
Freestyle9.64 runs/s
bochka26.69 runs/s

On the hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU / 8 GB, Vercel Sandbox did 12.96 runs per second and Freestyle did 9.64, so Vercel finished about a third more work in the same time. bochka did 26.69, about twice Vercel and 2.8 times Freestyle. This measures CPU speed only, not start time, where Freestyle advertises about 65 ms.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite.

Pricing in five scenarios

Each bill is priced from the providers' public price lists by the open Battleships engine, with the cheapest plan that runs the scenario.

ScenarioVercel Sandbox, a monthFreestyle, a monthbochka, a month
One coding agent, workdays
4 vCPU / 8 GB, 50 GB disk, 8 hours a day, 22 days a month, files kept between days
$57 · Pro$27 · Free$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$236 · Pro · with limits$159 · Free$26 · $20/mo
50 agents in parallel
50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside
$2,884 · Pro$2,139 · Pro$317 · $20/mo × 3 seats
Sandboxes for your users
2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once
$828 · Pro$581 · Pro$200 · $100/mo × 2 seats
1,000 short jobs at once
2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month
$3,152 · Pro$1,305 · Pro · with limits$2,000 · $2000/mo · with limits

Vercel Sandbox: Vercel charges for CPU only while your code uses it, but charges for memory for the whole time the sandbox runs, idle time included, plus a small fee per sandbox created and a monthly fee per GB of snapshots. The $20 Pro credit is shared with your other Vercel usage.

Freestyle: Freestyle bills per second for the vCPU, RAM and disk allocated to a VM, busy or idle, after monthly allowances of 200 vCPU-hours, 400 GiB-hours of RAM and 60,000 GiB-hours of storage. Hobby and Pro set a $50 or $500 monthly minimum that counts as usage, and a paused VM bills only for storage.

Freestyle: bochka charges a flat price per size, per second, only while the machine runs, and a stopped sandbox costs nothing and keeps its disk. Plans start at $20 a month, and all of the fee is usage credit.

Limits in those scenarios

For one agent that never stops, Vercel ends each session at 24 hours and running processes start over, while Freestyle runs on but has no idle timeout by default, so it bills until you pause it. For 1,000 short jobs at once, Vercel fits, but Freestyle Pro caps at 400 concurrent VMs. bochka does not fit that burst either: even the $2,000 plan allows 600 starts an hour and queues bursts above 90 starts a minute.

ScenarioVercel SandboxFreestylebochka
One coding agent, workdaysNone that biteNone that biteNone that bite
One agent that never stops
  • Pro: sessions capped at 24 h
None that biteNone that bite
50 agents in parallelNone that biteNone that biteNone that bite
Sandboxes for your usersNone that biteNone that biteNone that bite
1,000 short jobs at onceNone that bite
  • Pro: max 400 concurrent
  • shared vCPUs on this tier”): heavy, sustained CPU use can be slowed by other tenants
  • max 2,400 starts/day (needs ~6,667)
  • max 600 starts/hour (needs ~1,000)
  • bursts above 90 starts/min are queued (a burst of 333 takes ~4 min to start)

Which one should you use?

Use Vercel Sandbox when

you run many short bursts of untrusted code, need to run close to users around the world, or want to boot from your own Dockerfile.

  • Regions all over the world. Useful if your users are spread out: Vercel Sandbox runs in 19 regions, from Virginia to Tokyo and Sydney.
  • Huge bursts of short jobs. For many short runs at once, Pro allows far more parallel sandboxes than Freestyle Pro's 400, and CPU is billed only while used.
  • Images from your Dockerfile. Useful if you already have a Dockerfile: build it into an OCI image and boot sandboxes straight into it.

Use Freestyle when

your VMs must keep running or sleep between requests and resume with their memory, or when you need nested virtualization or machines bigger than 16 vCPU.

  • VMs that keep running. Useful for agents that live on one machine: a Freestyle VM has no session cap, while Vercel stops each session at 24 hours.
  • Memory pause, live fork and wake on request. Useful for VMs that sleep between requests: they resume the same processes and can wake on the next incoming traffic.
  • A full VM with nested virtualization. Useful for emulators or VMs inside: Freestyle runs systemd as pid 1 and supports KVM, which Firecracker does not offer.

Where bochka fits

If your agents work for hours on one machine, building, testing and checking what they built in a browser, look at bochka. It gives each one a full VM on CPUs about twice as fast as either, a desktop and browser built in, and an API to drive the agents inside from outside.

Faster builds and tests

bochka did 26.69 runs per second on the CPU benchmark, against 12.96 for Vercel Sandbox and 9.64 for Freestyle. Every install, build and test an agent waits on takes about two to three times as long on them.

A desktop and a browser, ready

Neither ships a screen, so an agent cannot open the app it built to see if it works. On Vercel you build the desktop yourself; on Freestyle you install GNOME or Xvfb, VNC and Chromium and wire up clicks yourself.

Agents you drive from outside

Both preinstall coding agent CLIs, but you start them as commands and parse their output yourself. bochka adds an API to prompt, steer and stream the agents inside, and can switch a credential to another account when one runs out.

Use bochka when your agents do real work for hours: builds, tests, Docker, emulators and a desktop with a browser, on fast CPUs, with an API to drive the agents inside from outside. Compare it one to one: bochka vs Vercel Sandbox and bochka vs Freestyle.

Questions people ask

Is Vercel Sandbox cheaper than Freestyle?

In our scenarios Freestyle costs less in four of five: one workday agent is $27 a month on Freestyle against $57 on Vercel, and 50 parallel agents are $2,139 against $2,884. For 1,000 short jobs at once, Vercel fits at $3,152 while Freestyle Pro hits its 400 VM cap. bochka is $20 a month for the workday agent and $317 for the 50 agents.

Which is faster, Vercel Sandbox or Freestyle?

On the hpc-sandbox-benchmarks CPU test at 4 vCPU / 8 GB, Vercel Sandbox did 12.96 runs per second and Freestyle did 9.64. bochka did 26.69 on the same test. Freestyle advertises VM starts in about 65 ms, which this test does not measure.

Can Vercel Sandbox or Freestyle run 24/7?

Freestyle can: a VM has no session cap and runs until you pause, stop or delete it. Vercel Sandbox sessions stop after 24 hours on Pro and Enterprise; the sandbox can resume from its snapshot, but processes and memory start fresh.

Can both run Docker and Android emulators?

Both run Docker. Only Freestyle offers nested virtualization, so Android emulators and VMs inside the sandbox work there but not on Vercel's Firecracker microVMs. bochka runs Docker and exposes /dev/kvm too.

Do Vercel Sandbox or Freestyle have a desktop for computer use?

No. Vercel Sandbox is headless only, and Freestyle has guides for installing a desktop, VNC and Chromium yourself. bochka includes a 60fps desktop, Chromium, and computer-use and browser-control APIs.

Where do they run?

Vercel Sandbox runs in 19 regions around the world, with iad1 in the US as the default. Freestyle does not list regions; its docs name a San Jose, US data center. bochka runs only in the EU (Germany, Finland and France).

Try bochka

Install the CLI, then run bochka new.

Product preview

More comparisons

Bochka product preview · Pricing, benchmarks and commands are illustrative. Sandbox provisioning is not connected.