Freestyle vs Agent 37
Freestyle gives every agent a full KVM VM with systemd and Docker that can pause with its memory and fork while running. Agent 37 starts with lighter gVisor sandboxes that sleep when idle, keep their memory and cost far less per hour, with a full VM as an opt-in type.
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.
Agent 37
Agent 37 hosts AI agents, one isolated, always-on sandbox per user, behind one API.
You create an instance from a template (Hermes, OpenClaw, Claude Code, Codex, OpenCode, Pi, Grok or your own Docker image) and talk to it through a chat API, files, exec, SSH and HTTPS URLs. The default instance is a container enforced by gVisor; a VM type gives a full Linux machine with its own kernel for Docker, Kubernetes and Android emulators. It is built for companies that give each of their customers an agent.
- Where it runs
- Dedicated servers in the United States
- Instance types
- Default (shared), Performance (dedicated cores), VM
- Compliance
- SOC 2 Type I, Type II in progress
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 workload | Freestyle: price rank | Agent 37: price rank | bochka: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Freestyle runs Docker and systemd on every VM; Agent 37 needs its VM type for Docker, which has no init system and bills compute at 2.5 times, and bochka runs both on faster CPUs. | 241st of 367, with limits $3,311 a month | 2nd cheapest of 367 $551 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Agent 37 ranks 1st here: one assistant per user that sleeps when idle, keeps its memory and wakes on the next message; Freestyle bills allocated vCPUs until you pause it. | 263rd of 366, with limits $4,195 a month | Cheapest of 366 $279 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 Both fit short code runs: Freestyle VMs pause with their memory and wake on the next request, and Agent 37's small default sandboxes cost little per minute. | 40th cheapest of 345 $482 a month | Cheapest of 345 $22 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 Freestyle Pro allows 400 VMs at once, while Agent 37 stops at 200 instances on self-serve, counting sleeping and stopped ones, so 300 users at once does not fit. | 16th cheapest of 361 $1,358 a month | 30th of 361, with limits $79 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; Agent 37 has no fork, so each rollout rebuilds its state from the start. | 107th of 360, with limits $6,600 a month | 6th of 360, with limits $363 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 | 90th of 364, with limits $58 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 Freestyle's docs send CPU-heavy jobs like kernel builds elsewhere, Agent 37 has no published CPU score, and bochka did 2.8 times Freestyle's work on the CPU benchmark. | 168th of 377, with limits $625 a month | 2nd cheapest of 377 $84 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB | 98th cheapest of 368 $151 a month | 3rd cheapest of 368 $12 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 Agent 37 adds a noVNC desktop through a template build, on Freestyle you install the desktop, VNC and input tools yourself, and bochka streams a 60fps desktop with computer use built in. | 234th of 369, with limits $940 a month | Cheapest of 369 $53 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 | 48th of 367, with limits $29 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 | Cheapest of 346 $32 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 | 36th of 338, with limits $648 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 Agent 37 ranks 1st on cost here, but it has no fork; Freestyle can clone a running VM with its memory, and bochka can fork a running sandbox to try many paths at once. | 272nd of 360, with limits $482 a month | Cheapest of 360 $43 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 it | Can't run it | Can't run it |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Can't run it | Can't run it | Can't run it |
Key features
| Feature | Freestyle | Agent 37 | bochka |
|---|---|---|---|
| Isolation | Full KVM VM with its own kernel for every sandbox | gVisor by default; full VM with its own kernel as an opt-in type | Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts |
| Root, systemd and Docker | Yes; systemd is pid 1 and Docker runs from boot | Docker only on the VM type, at 2.5x the default compute rate; no init system on any type | Yes, in every sandbox, with systemd as on any Linux server |
| Nested virtualization | Full KVM support | /dev/kvm on the VM type, with a ready Android template | /dev/kvm in every sandbox, Android emulator works |
| CPU benchmark (4 vCPU / 8 GB) | 9.64 runs/s | No published result; Performance type on dedicated cores at 4x the compute rate | 26.69 runs/s |
| Biggest machine | Up to 32 vCPU / 64 GiB on Pro, 64 / 128 with custom limits | Up to 16 vCPU / 32 GB after a first top-up | Up to 16 vCPU / 32 GB on the $100 plan and above |
| Sleep and wake with memory | Pause saves memory and disk; a VM paused by idle timeout wakes on the next traffic | Opt-in auto-sleep after a quiet period (default 15 minutes), memory checkpointed, wakes on the next request | No idle timer; disk is saved every minute and on stop, and a resume boots the machine again |
| Fork and snapshots | Clone a running VM without pausing it; snapshots capture memory and disk | No fork; daily backups plus one on-demand backup, restored in place only | Fork a running sandbox with its disk; named snapshots as shareable templates |
| Coding agents | claude, codex, opencode and openclaw preinstalled, driven through exec or a terminal | Hermes, OpenClaw, Claude Code, Codex, Grok, OpenCode and Pi templates, with a chat API per instance | Seven agents preinstalled, with a prompt, steer and events API |
| Desktop, browser and computer use | Guides to install GNOME or Xvfb, VNC and Chromium yourself | noVNC desktop through a template build or the Ubuntu VM template | Linux desktop streamed at 60fps, Chromium and computer use built in |
| Keeping API keys out of the sandbox | Credentials injected at the edge for HTTPS headers, Postgres and S3; the VM must trust the Freestyle CA | Managed LLM, search and Composio through a per-instance token; your own keys go in as env vars | Proxy for model providers and any HTTPS API; switch a credential to another account when credits run out |
| Outbound rules | Default-deny firewall with allow rules by IP, VM, VPC or exact domain | Open internet, no outbound allowlist | Open internet, no provider-side allowlist |
| Inbound access | HTTPS on free style.dev names or your own domain, forward auth, other public protocols | HTTPS URL per port, signed URLs, custom domains; no public IP, SSH through a WebSocket tunnel | HTTPS URLs for up to 50 ports, token-protected by default; raw TCP |
| Idle billing | A paused VM bills for its storage; no idle timeout unless you set one | Disk billed at $0.09 per GB a month even while stopped or asleep | A stopped sandbox costs nothing and keeps its disk |
| Regions | Not listed; the docs name a San Jose, US data center | Dedicated servers in the United States | EU only (Germany, Finland, France), EU data residency |
Every feature Battleships tracks (43)
| Feature | Freestyle | Agent 37 | bochka |
|---|---|---|---|
| Full VM (own kernel) | Yes | No | Yes |
| Docker inside | Yes | No | Yes |
| Nested virtualization | Yes | No | Yes |
| Root | Yes | Yes | Yes |
| Systemd | Yes | Not published | Yes |
| Desktop GUI | No | Yes | Yes |
| Browser | No | Yes | Yes |
| Computer-use API | No | Yes | Yes |
| SSH | Yes | Yes | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | Yes | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Yes | Yes | Yes |
| Memory fork | Yes | No | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | Yes | No |
| Wake on request | Yes | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | Yes | Not published | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | Yes | No | No |
| Secret proxy | Yes | Yes | Yes |
| Secret proxy for any API | Yes | No | Yes |
| Preinstalled agents | Yes | Yes | Yes |
| Agent harness API | No | Yes | Yes |
| MCP server | No | No | No |
| GPU | Yes | No | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | No | No | No |
| BYOC | No | No | No |
| Open source | No | No | No |
| Scoped API keys | Yes | Not published | Yes |
| Spend limits | Not published | Yes | Yes |
| Usage per sandbox (API) | No | Yes | Yes |
| Audit logs | Yes | Not published | No |
| SSO | Yes | No | No |
| SOC 2 | Not published | Yes | No |
| HIPAA | Not published | No | No |
| EU data residency | No | No | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
On the hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU / 8 GB, Freestyle did 9.64 runs per second and bochka did 26.69, about 2.8 times the work in the same time. Agent 37 has no published result on that benchmark, so there is no comparable measurement. Its default instances share host resources, and its Performance type runs on dedicated cores at 4x the compute rate.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Agent 37 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.
| Scenario | Freestyle, a month | Agent 37, a month | bochka, 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 | $6 · Wallet (any top-up) | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $159 · Free | $12 · Wallet (any top-up) | $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 | $329 · Wallet ($100 lifetime top-ups) | $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 | $33 · Wallet ($250 lifetime top-ups) · with limits | $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 | $73 · Wallet ($250 lifetime top-ups) · with limits | $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 set a $50 or $500 monthly minimum that counts as usage.
Agent 37: Agent 37 uses a prepaid wallet with no plan fee, metered per minute: $0.80 per vCPU and $0.70 per GB of RAM per month while running, and $0.09 per GB of disk per month even while stopped or asleep. Performance compute costs 4x and VM compute 2.5x the default rate.
Agent 37: 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 Pro allows 400 VMs at once, which is why it does not fit 1,000 short jobs at once, and on Free the monthly allowances are hard stops. Agent 37's instance cap is 10 after a first top-up, 50 at $100 of total top-ups and 200 at $250, and it counts sleeping and stopped instances, so neither 300 user sandboxes at once nor 1,000 jobs fit. Agent 37 does not publish an egress price, and its 50-agent fleet is priced on the VM type because it needs Docker. bochka's limit in the burst is starts: its top plan queues starts above 90 a minute.
| Scenario | Freestyle | Agent 37 | bochka |
|---|---|---|---|
| One coding agent, workdays | None that bite |
| None that bite |
| One agent that never stops | None that bite |
| None that bite |
| 50 agents in parallel | None that bite |
| None that bite |
| Sandboxes for your users | None that bite |
| None that bite |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use Freestyle when
you want a full VM per agent that pauses with its memory, forks while running, sits behind a default-deny firewall, or needs more than 16 vCPU.
- A full VM by default. Every Freestyle VM has its own kernel, systemd and Docker running, which Agent 37 only offers on its pricier VM type.
- Fork a running VM with its memory. For work that branches from one state: clone a running VM without pausing it, or boot copies from a memory snapshot.
- Network controls and bigger machines. Each VM starts behind a default-deny firewall, and Pro machines reach 32 vCPU / 64 GiB, twice Agent 37's largest shape.
Use Agent 37 when
you host one mostly idle assistant per customer in the US that should sleep with its memory kept, and low per-minute rates matter more than Docker or forks.
- Sleeps when idle and keeps its memory. For assistants that sit idle most of the day: an instance sleeps, keeps its memory and wakes on the next message.
- Low per-minute rates, no plan fee. A prepaid wallet metered per minute, so one always-on agent costs $12 a month in the table, against $159 on Freestyle.
- Agent templates and your own image. Start from Hermes, OpenClaw, Claude Code and others with a chat API, or from your own Docker image.
Where bochka fits
If your agents spend hours on installs, builds, tests and Docker, or need a desktop and a browser, look at bochka. It gives every sandbox a full VM like Freestyle, on faster CPUs, with a desktop and an agent API built in, and a stopped sandbox costs nothing.
A full VM on fast CPUs, by default
Agent 37's default gVisor sandbox cannot run Docker, and its VM type bills 2.5 times the compute with no systemd. Freestyle gives a full VM but did 9.64 runs per second to bochka's 26.69, so every build waits longer.
A desktop and a browser, ready
On Freestyle you install GNOME or Xvfb, VNC and Chromium yourself and build the screenshot and click layer. On Agent 37 a noVNC desktop needs a template build. bochka streams a 60fps desktop with computer use in every sandbox.
Agents and keys driven from outside
On Agent 37 your own API keys sit as env vars inside a machine the agent controls. On Freestyle you drive agent CLIs through exec and parse their output. bochka proxies any HTTPS API and gives you a prompt, steer and events API.
Use bochka when your agents do real work for hours: builds, tests, Docker, emulators, a desktop and a browser, on fast CPUs, with an API to drive the agents inside from outside. Compare it one to one: bochka vs Freestyle and bochka vs Agent 37.
Questions people ask
Is Agent 37 cheaper than Freestyle?
Yes, in every scenario on this page. One agent running all month is $12 on Agent 37 against $159 on Freestyle, and 50 agents in parallel are $329 against $2,139, even on Agent 37's VM type. Agent 37 still does not fit 300 user sandboxes at once, because its cap is 200 instances.
Which is faster, Freestyle or Agent 37?
Freestyle did 9.64 runs per second on the hpc-sandbox-benchmarks CPU test at 4 vCPU / 8 GB; Agent 37 has no published result, so they cannot be compared. bochka did 26.69 on the same test. Freestyle also advertises VM starts in about 65 ms, which this test does not measure.
Can Agent 37 run Docker like Freestyle?
Only on its VM type, which bills compute at 2.5 times the default rate; the default gVisor type cannot run a Docker daemon. Freestyle runs Docker from boot on every VM, and so does bochka.
Do both keep memory when a sandbox sleeps?
Yes. Freestyle pauses a VM with its memory and can wake it on the next traffic, and Agent 37 checkpoints memory on auto-sleep and wakes on the next request. bochka keeps the disk but boots the machine again on resume.
Where do Freestyle and Agent 37 run?
Agent 37 runs on dedicated servers in the United States. Freestyle does not list regions, and its docs name a San Jose, US data center. bochka runs only in the EU, in Germany, Finland and France.
Try bochka
Install the CLI, then run bochka new.
Product preview