Freestyle vs Beam

Freestyle gives each agent a full KVM virtual machine with systemd and Docker that can pause with its memory and wake on the next request. Beam is a serverless GPU platform whose sandboxes are quick gVisor containers, with a separate line of CPU VMs.

The companies and products

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.

Beam

Beam is a serverless cloud for GPU inference, sandboxes and background jobs, built on its open-source runtime Beta9.

Beam runs functions, REST endpoints, task queues, pods, sandboxes and VMs on CPUs and GPUs, all from a Python or TypeScript SDK. A sandbox is a container isolated with gVisor that starts in a second or two, lives until you terminate it or its time to live runs out, and can be saved as a filesystem image or a memory snapshot. Beam VMs are a separate CPU product: a microVM with its own kernel, systemd and a root disk that stays when it stops. Much of Beam is a GPU serverless platform for AI teams.

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. “40th cheapest of 345” means 39 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 workloadFreestyle: price rankBeam: price rankbochka: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Freestyle and bochka both give a full VM with Docker, systemd and /dev/kvm, while a Beam sandbox is a gVisor container without systemd; bochka's CPUs did about 2.8 times Freestyle's work.
241st of 367, with limits
$3,311 a month
69th of 367, with limits
$551 a month
3rd cheapest of 367
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
263rd of 366, with limits
$4,195 a month
115th of 366, with limits
$651 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
Beam fits short code runs best of the three: sandboxes start in a second or two, bill per second and go away when their timer ends.
40th cheapest of 345
$482 a month
8th cheapest of 345
$117 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
All three fit; Beam costs less than Freestyle here, Freestyle can wake a paused user VM on its next request, and bochka gives each user a full VM with hosting.
16th cheapest of 361
$1,358 a month
3rd cheapest of 361
$290 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
Freestyle can snapshot a running VM with its memory and boot copies from it; Beam can restore a memory snapshot into new sandboxes too.
107th of 360, with limits
$6,600 a month
341st of 360, with limits
$939 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
53rd cheapest of 364
$1,046 a month
13th cheapest of 364
$251 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
Beam runs CI jobs for much less than Freestyle, but jobs that need /dev/kvm for Android emulators or VM images need Freestyle or bochka.
168th of 377, with limits
$625 a month
12th cheapest of 377
$171 a month
4th cheapest of 377
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A machine you keep all month fits a Beam VM, which keeps a 50 GiB root disk while stopped, or a bochka sandbox, which keeps its disk for free.
98th cheapest of 368
$151 a month
13th cheapest of 368
$21 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: Freestyle has guides to install one yourself and Beam has no desktop product, while bochka streams a 60fps desktop with computer use built in.
234th of 369, with limits
$940 a month
111th of 369, with limits
$237 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
257th of 367, with limits
$482 a month
195th of 367, with limits
$176 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
278th of 346, with limits
$482 a month
200th of 346, with limits
$131 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
183rd of 338, with limits
$6,560 a month
42nd of 338, with limits
$923 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
272nd of 360, with limits
$482 a month
175th of 360, with limits
$155 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
Beam has GPU sandboxes and serverless GPUs up to H100, H200 and B200; Freestyle offers private GPU sandboxes only on Enterprise, and bochka is CPU only.
Can't run itCheapest of 98
$458 a month
Can'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

FeatureFreestyleBeambochka
IsolationFull KVM VM with its own kernel per sandboxgVisor container by default; separate Beam VMs are CPU microVMs with their own kernelFull VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
CPU benchmark (4 vCPU / 8 GB)9.64 runs/sNo result on this test26.69 runs/s
Root, systemd and DockerYes; systemd is pid 1 and Docker runs from bootDocker in sandboxes with docker_enabled; systemd only in Beam VMsYes, like a normal Ubuntu server
Nested virtualizationYes, full KVM supportNot documentedYes, /dev/kvm for Android emulators and VMs
Biggest machine32 vCPU / 64 GiB on Pro, 64 / 128 with custom limitsMax sandbox size not published16 vCPU / 32 GB on the $100 plan and above
Memory pause, fork and wakePauses with memory, clones a running VM, wakes a paused VM on the next requestMemory snapshot of a running sandbox, restored into new sandboxes; no wake on requestForks a running sandbox with its disk; disk saved every minute, resume boots again
Idle billingNo idle timeout by default; a paused VM bills for its storageA sandbox bills until terminated or its time to live ends; a stopped Beam VM has no compute chargeA stopped sandbox costs nothing and keeps its disk
GPUsPrivate GPU sandboxes on EnterpriseGPU sandboxes and serverless GPUs up to H100, H200 and B200No, CPU only
Desktop, browser and computer useGuides to install GNOME or Xvfb, VNC and Chromium yourselfNo desktop or browser productLinux desktop streamed at 60fps, Chromium and computer use built in
Coding agentsclaude, codex, opencode and openclaw preinstalled, run through exec or a terminalNone preinstalled; you install and run them through process execSeven agents preinstalled, with a prompt, steer and events API
Keeping API keys out of the sandboxCredentials injected at the edge for HTTPS headers, Postgres and S3Secrets are environment variables inside the sandboxCredentials proxy for model providers and any HTTPS API; switch to another account when credits run out
Inbound accessHTTPS on free style.dev names or your own domain, forward auth, other public protocolsPublic HTTPS URL per exposed port with no built-in auth; no raw TCPHTTPS URLs for up to 50 ports, token-protected by default; raw TCP
Outbound rulesDefault-deny firewall with allow rules by IP, VM, VPC or exact domainBlock all outbound traffic, or allow up to 10 CIDR rangesOpen internet, no provider-side allowlist
RegionsNot listed; the docs name a San Jose, US data centerUS only, on AWS and Google CloudEU only (Germany, Finland, France)
Every feature Battleships tracks (43)
FeatureFreestyleBeambochka
Full VM (own kernel)YesNoYes
Docker insideYesYesYes
Nested virtualizationYesNoYes
RootYesYesYes
SystemdYesNoYes
Desktop GUINoNoYes
BrowserNoNoYes
Computer-use APINoNoYes
SSHYesYesYes
Persistent diskYesYesYes
Pause/resumeYesNoYes
Automatic snapshotsYesNoYes
Fork/cloneYesYesYes
Live fork (no pause)YesYesYes
Memory forkYesYesNo
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesYesYes
Idle auto-stopYesNoNo
Wake on requestYesNoNo
HTTPS preview URLsYesYesYes
Raw TCP inboundYesNoYes
Public IPv4NoNoNo
Egress allowlistYesYesNo
Secret proxyYesNoYes
Secret proxy for any APIYesNoYes
Preinstalled agentsYesNoYes
Agent harness APINoNoYes
MCP serverNoYesNo
GPUYesYesNo
WindowsNoNoNo
Arm64NoNoNo
Self-hostingNoYesNo
BYOCNoYesNo
Open sourceNoYesNo
Scoped API keysYesNot publishedYes
Spend limitsNot publishedNot publishedYes
Usage per sandbox (API)NoNoYes
Audit logsYesNot publishedNo
SSOYesNoNo
SOC 2Not publishedYesNo
HIPAANot publishedYesNo
EU data residencyNoNot publishedYes
Zero data retentionNot publishedNot publishedYes

Performance

Freestyle9.64 runs/s
BeamNot measured
bochka26.69 runs/s

On the hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU / 8 GB, bochka did 26.69 runs per second and Freestyle did 9.64, so bochka did about 2.8 times the CPU work. Beam has no result on this test, so we cannot say how its CPUs compare. This test measures CPU speed only: Freestyle advertises VM starts in about 65 ms, and Beam's guide says its sandboxes cold-boot in 1 to 3 seconds.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Beam has no published run in this suite yet, so there is no like-for-like number to compare.

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.

ScenarioFreestyle, a monthBeam, 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
$27 · Free$5 · Developer$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$159 · Free$22 · Developer$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
$2,139 · Pro$410 · Team$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
$581 · Pro$191 · Team$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
$1,305 · Pro · with limits$259 · Team$2,000 · $2000/mo · with limits

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. On Free those allowances are hard stops; Hobby and Pro have a $50 or $500 monthly minimum that counts as usage.

Beam: Beam sandboxes bill per second for each CPU core (two vCPU) and GiB of RAM until you terminate them or their time to live ends, so an idle tail is billed too. Developer has no fee; Team adds $89 a month with no usage credit, and there are no egress fees.

Beam: Plans start at $20 a month, and the whole fee is usage credit. You pay per second while a sandbox runs, and a stopped sandbox costs nothing and keeps its disk.

Limits in those scenarios

Freestyle's Free allowances are hard stops, and its Pro plan allows 400 concurrent VMs, which is why it does not fit 1,000 short jobs at once. Beam's Developer plan allows 30 CPU containers at once, so 50 parallel agents need the Team plan, which allows 1,000. Beam publishes no price for disk beyond what a VM includes, so its scenarios count no disk cost. bochka's start limits are lower: its $2,000 plan queues starts above 90 a minute.

ScenarioFreestyleBeambochka
One coding agent, workdaysNone that bite
  • disk beyond 0 GiB: price unknown
None that bite
One agent that never stopsNone that bite
  • disk beyond 0 GiB: price unknown
None that bite
50 agents in parallelNone that bite
  • disk beyond 0 GiB: price unknown
None that bite
Sandboxes for your usersNone that bite
  • disk beyond 0 GiB: price unknown
None that bite
1,000 short jobs at once
  • Pro: max 400 concurrent
  • shared vCPUs on this tier”): heavy, sustained CPU use can be slowed by other tenants
  • disk beyond 0 GiB: price unknown
  • 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 Freestyle when

your agents need a full VM with systemd and nested virtualization, when VMs must sleep with their memory and wake on the next request, or when you need domain-level outbound rules.

  • A real VM, not a container. For agents that need systemd, Docker from boot or nested virtualization, every Freestyle sandbox is a full KVM VM.
  • Pause with memory, wake on request. For VMs that sleep between requests, Freestyle resumes the same processes and can wake on the next incoming traffic.
  • Domain-level network rules. Each VM starts with no network access, and you allow exactly the IPs, VMs, VPCs or domains it may reach.

Use Beam when

your agents need GPUs, when you run many short code runs and want the lowest bill, or when you want to self-host the open-source runtime.

  • GPUs in the same platform. For work that needs a GPU, Beam offers GPU sandboxes and serverless GPUs from RTX 4090 up to H100, H200 and B200.
  • Lower bills and big bursts. Beam costs less than Freestyle in every scenario here, and its Team plan fits 1,000 short jobs at once.
  • Open source and your own cloud. Its Beta9 runtime is AGPL-3.0 and free to self-host, and it can run in your own cloud for a management fee.

Where bochka fits

If your agents do real work for hours, like installs, builds, Docker, emulators and a browser, look at bochka. Each sandbox is a full VM on fast CPUs with a desktop, coding agents and a credentials proxy built in, and a stopped sandbox costs nothing.

A full VM on fast CPUs

A Beam sandbox is a gVisor container with no systemd and no documented /dev/kvm. Freestyle is a full VM, but it did 9.64 runs/s on the CPU test against bochka's 26.69, so every build waits about 2.8 times longer.

A desktop and browser, ready

Beam has no desktop or browser product, and on Freestyle you install GNOME or Xvfb, VNC and Chromium into your own snapshot, then build the screenshot and click layer yourself before an agent can use a web page.

Agents and keys driven from outside

On Freestyle you run agent CLIs through exec and parse the output; on Beam you install them yourself and secrets sit as environment variables the agent can read. bochka has a prompt and steer API and keeps real keys in a proxy.

Use bochka when your agents do real work for hours on a full VM with fast CPUs, Docker, /dev/kvm, a 60fps desktop and a browser, with an API to drive the agents inside from outside. Compare it one to one: bochka vs Freestyle and bochka vs Beam.

Questions people ask

Is Freestyle cheaper than Beam?

No. Beam costs less in all five scenarios on this page, such as $22 against $159 a month for one agent that never stops, and $410 against $2,139 for 50 agents in parallel. Beam publishes no price for extra disk, so its numbers count none.

Which is faster, Freestyle or Beam?

We cannot say for CPU work: Freestyle did 9.64 runs per second on the hpc-sandbox-benchmarks test at 4 vCPU / 8 GB, and Beam has no result on it. bochka did 26.69 on the same test. For starts, Freestyle advertises about 65 ms and Beam says 1 to 3 seconds.

Can Beam and Freestyle run Docker?

Yes. Freestyle runs Docker from boot under systemd in a full VM, and Beam supports Docker inside a sandbox with docker_enabled. Only Freestyle documents nested virtualization for Android emulators and VMs, and bochka has /dev/kvm too.

Is a Beam sandbox a VM like Freestyle's?

Not by default. Beam sandboxes are containers using runc and gVisor, with an opt-in microVM mode for CPU sandboxes, and Beam sells separate CPU VMs with systemd. Every Freestyle VM and every bochka sandbox is a full VM with its own kernel.

Which one has GPUs?

Beam does, with GPU sandboxes and serverless GPUs from RTX 4090 up to H100, H200 and B200. Freestyle lists private GPU sandboxes only for Enterprise customers, and bochka is CPU only.

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.