boat vs Google Compute Engine
Google Compute Engine is a hyperscaler's general-purpose VM service: you get a machine and build the rest. boat is a sandbox made for agents, where every machine already has Docker, /dev/kvm, a desktop, coding agents and HTTPS URLs, and you can fork it while it runs.
Where boat stands out
Coding agents you drive from outside
boat ships Claude Code, Codex, OpenCode and other agents inside each sandbox, and you drive them from outside with a prompt, steer and events API. Its credentials proxy keeps real AI keys out of the sandbox and can switch to another account when one runs out of credits. Compute Engine gives you a clean VM and a shell.
/dev/kvm, minute snapshots and forks on every machine
Every boat sandbox exposes /dev/kvm for Android emulators and VMs, snapshots its disk every minute and on stop, and can be forked while it runs. A stopped sandbox costs nothing. On Compute Engine, nested virtualization is off for E2, Arm and most AMD VMs, and a copy means making a machine image and booting a new VM.
Desktop, browser and HTTPS URLs ready
Every boat sandbox has a 60fps desktop with Chromium and computer use, and HTTPS URLs on up to 50 ports, token-protected. Compute Engine Linux images have no desktop, and an HTTPS URL needs your own certificate plus a load balancer or web server.
The companies and products
boat
boat.dev runs sandboxes for AI agents: each one is a full Linux VM with its own kernel.
You start a machine with one command or one API call, then SSH in, run Docker, open a desktop, or let a built-in agent (Claude Code, Codex, OpenCode and others) work in it. The disk stays between sessions, and a stopped sandbox costs nothing.
- Isolation
- Full VM per sandbox, own kernel
- Regions
- EU: Germany, Finland, France
Google Compute Engine
Google Compute Engine is Google Cloud's general-purpose VM service, not a sandbox product: you rent virtual machines by the second and set them up yourself.
You pick a machine type (E2, N2D, C3D, Arm T2A, GPU families and more), a boot image and a network, and Compute Engine boots a KVM-based VM with its own kernel and root access. Disks are Persistent Disk or Hyperdisk volumes, traffic goes through VPC firewall rules, and anything agent-specific (agents, desktops, HTTPS URLs, key handling) is yours to build. It is built for every kind of server workload, from websites to GPU training, across Google Cloud's worldwide regions.
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. “3rd cheapest of 367” means 2 of the other 366 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 | boat: price rank | Google Compute Engine: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Compute Engine can run a software factory, but you build the agent setup, preview URLs and cloning yourself; boat ships them on every machine. | 3rd cheapest of 367 $634 a month | 120th of 367, with limits $1,106 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An always-on assistant runs on both; on Compute Engine the cheapest always-on price needs a 3-year commitment, and its disk bills even while stopped. | 5th cheapest of 366 $657 a month | 177th of 366, with limits $1,433 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Compute Engine handles big bursts of short jobs better than boat's start limits allow, though each VM takes a full boot to start. | 186th of 345, with limits $2,000 a month | 92nd of 345, with limits $133 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month On Compute Engine you build HTTPS access, isolation per user and lifecycle tooling yourself; boat ships them for each sandbox. | Cheapest of 361 $209 a month | 92nd of 361, with limits $836 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 34th of 360, with limits $2,000 a month | 36th of 360, with limits $2,085 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 10th cheapest of 364 $200 a month | 16th cheapest of 364 $263 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month CI that builds Android apps or VM images needs /dev/kvm, which boat has on every machine and Compute Engine's E2 VMs do not. | 4th cheapest of 377 $86 a month | 41st cheapest of 377 $359 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB | 18th cheapest of 368 $26 a month | 82nd cheapest of 368 $109 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Compute Engine Linux images have no desktop or computer-use API, so you install and stream one yourself; boat streams a 60fps desktop out of the box. | 6th cheapest of 369 $200 a month | 129th of 369, with limits $293 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month boat has Chromium and browser control ready in each sandbox; on Compute Engine you install and wire the browser yourself. | 7th cheapest of 367 $200 a month | 212th of 367, with limits $244 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 6th cheapest of 346 $90 a month | 218th of 346, with limits $179 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 71st of 338, with limits $1,800 a month | 108th of 338, with limits $2,978 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month | 2nd cheapest of 360 $200 a month | 152nd of 360, with limits $120 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month Only Compute Engine fits here: it rents Nvidia GPUs such as L4, H100, H200 and B200, and boat has no GPUs. | 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 |
Key features
| Feature | boat | Google Compute Engine |
|---|---|---|
| What you get | An agent sandbox: a full VM with agents, desktop, URLs and snapshots set up | A general cloud VM; you install and wire everything else |
| Isolation | Every sandbox is a full VM with its own kernel | Every instance is a full KVM-based VM with its own kernel |
| Docker and nested virtualization | Docker, root, systemd and /dev/kvm on every machine | Docker, root and systemd on any VM; nested KVM only on some Intel families and N4D, not on E2, Arm or other AMD VMs |
| Coding agents | Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API | None preinstalled; you install agents and drive them over SSH or startup scripts |
| Keeping keys out of the machine | AI credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out | No credential proxy; service account tokens and Secret Manager values are read from inside the VM |
| Desktop and computer use | 60fps desktop and Chromium, computer use and browser control, embeddable stream | No managed desktop, browser or computer-use API on Linux VMs |
| Reaching the machine | HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4 | External IPv4 and any port through firewall rules; HTTPS needs your own certificate and a load balancer or web server |
| Files after a stop | Snapshots every minute and on stop; a stopped sandbox costs nothing | Persistent disks are kept and billed while stopped; snapshots are on demand or on a schedule you set |
| Pause and fork | Fork a running sandbox; resume reboots from disk (no memory snapshots) | Suspend keeps memory for the same VM for up to 60 days; cloning means making a machine image (disks, not memory) and booting a new VM |
| Outbound network rules | Open internet on every plan; no provider-side allowlist | VPC egress firewall rules by IP and port; domain (FQDN) rules on Cloud NGFW Standard, a paid tier |
| GPUs | None | Nvidia GPUs from T4 and L4 up to H100, H200, B200 and GB200 |
| Operating systems and CPUs | Linux on AMD Ryzen 9 9950X (Zen 5); Wine for Windows apps | Linux and Windows Server on Intel, AMD and Arm; no macOS |
| Regions | EU only: Germany, Finland, France | 43 regions worldwide, including many in the EU |
| Compliance and transparency | SOC 2 is in progress. We're open about where we stand: live status and data retention policy. | SOC 2 |
Every feature Battleships tracks (43)
| Feature | boat | Google Compute Engine |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | Yes |
| Nested virtualization | Yes | No |
| Root | Yes | Yes |
| Systemd | Yes | Yes |
| Desktop GUI | Yes | No |
| Browser | Yes | No |
| Computer-use API | Yes | No |
| SSH | Yes | Yes |
| Persistent disk | Yes | Yes |
| Pause/resume | Yes | Yes |
| Automatic snapshots | Yes | No |
| Fork/clone | Yes | Yes |
| Live fork (no pause) | Yes | Yes |
| Memory fork | No | No |
| Custom image (Docker or snapshot) | Yes | Yes |
| ≥24 h sessions | Yes | Yes |
| Idle auto-stop | No | No |
| Wake on request | No | No |
| HTTPS preview URLs | Yes | No |
| Raw TCP inbound | Yes | Yes |
| Public IPv4 | No | Yes |
| Egress allowlist | No | Yes |
| Secret proxy | Yes | No |
| Secret proxy for any API | Yes | No |
| Preinstalled agents | Yes | No |
| Agent harness API | Yes | No |
| MCP server | No | No |
| GPU | No | Yes |
| Windows | No | No |
| Arm64 | No | Yes |
| Self-hosting | No | No |
| BYOC | No | No |
| Open source | No | No |
| Scoped API keys | Yes | Not published |
| Spend limits | Yes | No |
| Usage per sandbox (API) | Yes | Yes |
| Audit logs | No | Yes |
| SSO | No | Yes |
| SOC 2 | No | Yes |
| HIPAA | No | Yes |
| EU data residency | Yes | Yes |
| Zero data retention | Yes | Not published |
Performance
There is no comparable measurement: Compute Engine was not part of the public benchmark run we use. In that run, StarSling's Node.js build-and-test test on 2026-09-29, boat did 26.69 runs per second at 4 vCPU / 8 GB. Compute Engine speed depends on the machine family you pick, from E2 to the latest Intel and AMD chips.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Google Compute Engine 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 | boat, a month | Google Compute Engine, 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 | $20 · $20/mo | $31 · Pay-as-you-go billing account (no plan fee) |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $64 · Pay-as-you-go billing account (no plan fee) · with limits |
| 50 agents in parallel 50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside | $317 · $20/mo × 3 seats | $1,291 · Pay-as-you-go billing account (no plan fee) |
| Sandboxes for your users 2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once | $200 · $100/mo × 2 seats | $378 · Pay-as-you-go billing account (no plan fee) |
| 1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month | $2,000 · $2000/mo · with limits | $690 · Pay-as-you-go billing account (no plan fee) |
boat: You pick a monthly plan from $20, and all of it becomes usage credit spent per second on running sandboxes. A stopped sandbox costs nothing and keeps its disk.
Google Compute Engine: Compute Engine bills each VM per second after a 1 minute minimum at its list rate, here E2 sizes on demand in us-central1, with no platform fee. Disks and external IPs bill on top, also while stopped, and lower rates need a 1-year or 3-year commitment that bills every hour of the term.
Limits in those scenarios
Compute Engine limits vCPUs per region with quotas that vary by project, and some new projects also have a global CPU cap, so 50 agents in parallel or 300 sandboxes for your users may need a quota request first. A free trial account cannot ask for more. One agent that never stops reaches the price shown only with a 3-year commitment. On boat, the 1,000-job burst is the one that does not fit: the $2,000 plan allows 90 starts a minute, 600 an hour and 2,400 a day, so the burst is queued.
| Scenario | boat | Google Compute Engine |
|---|---|---|
| One coding agent, workdays | None that bite | None that bite |
| One agent that never stops | 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 |
| None that bite |
Which one should you use?
Use Google Compute Engine when
you need GPUs, Windows, regions outside Europe, or very large bursts of short jobs, and you have the people to build the agent tooling on top.
- GPUs, Windows and Arm. Useful if your agents need GPUs or Windows: Compute Engine rents Nvidia GPUs up to B200 and runs Windows Server.
- Regions all over the world. Useful if your users or data must sit outside Europe: Compute Engine runs in 43 regions and 130 zones.
- Network rules and memory suspend. Useful if you must limit where code can connect, or resume a VM with its memory intact.
Use boat when
you want each agent on a ready, full VM: coding agents with an API, keys kept outside, /dev/kvm, a desktop, HTTPS URLs, minute-by-minute snapshots and forks, with no infrastructure to assemble.
- Coding agents you drive from outside. On Compute Engine you write the code that installs, starts, prompts and watches every agent, and keys sit inside the VM where any root process can read them. That is platform work before the first task runs.
- /dev/kvm, minute snapshots and forks on every machine. An agent that builds Android apps cannot start an emulator on an E2 VM. Each branch on Compute Engine waits for an image and a fresh boot, and every parked VM keeps billing its disk.
- Desktop, browser and HTTPS URLs ready. A computer-use agent on Compute Engine needs you to install a desktop, a streaming server and a click API. Showing a user the app an agent built needs DNS and TLS work for every VM.
Questions people ask
Is boat cheaper than Google Compute Engine?
In four of the five scenarios we priced, yes. One coding agent on workdays costs $20 a month on boat and $31 on Compute Engine, 50 agents in parallel come to $317 on boat versus $1,291, and sandboxes for your users to $200 versus $378. Compute Engine wins the 1,000-job burst at $690 a month, which does not fit boat's start limits.
Can I just run Claude Code on a Compute Engine VM?
Yes. You create a VM, install the agent, give it a key and keep it running over SSH. boat does that work for you: the agents come preinstalled, you drive them with a prompt, steer and events API, and the real keys stay outside the machine.
Does Compute Engine support nested virtualization like boat?
On some machine types. Google's docs say nested virtualization works only with Linux KVM, must be enabled per VM, and is not available on E2, memory-optimized, Arm or AMD VMs except N4D. On boat, /dev/kvm is available on every machine.
Does boat have GPUs or Windows like Compute Engine?
No. boat runs Linux VMs on dedicated AMD Ryzen 9 9950X (Zen 5) hosts, with no GPUs and Wine docs for Windows apps. If you need Nvidia GPUs or Windows Server, Compute Engine has both.
Where does my data run?
boat runs only in the EU (Germany, Finland and France), with EU data residency and a zero data retention option. Compute Engine runs in 43 regions worldwide, including many in the EU, and you choose the region for each VM.
Is boat SOC 2 compliant?
SOC 2 is in progress. We're open about where we stand: live status at status.ascii.dev and our data retention policy at docs.boat.dev/data-retention.
Start a sandbox
Install the CLI, then run boat new.
7-day free trial
More comparisons
- boat vs Oracle Cloud
- boat vs IONOS Cloud
- boat vs AWS AgentCore
- boat vs Contabo
- boat vs Azure Virtual Machines
- boat vs AWS Fargate
- boat vs netcup
- boat vs AWS EC2
- boat vs E2B
- boat vs Daytona
- boat vs Modal
- boat vs Vercel Sandbox
- AWS EC2 vs Google Compute Engine
- Google Compute Engine vs Azure Virtual Machines
- Google Compute Engine vs Oracle Cloud
- Google Compute Engine vs DigitalOcean
- Google Compute Engine vs Hetzner Cloud
- Google Compute Engine vs Vultr
- Google Compute Engine vs Linode
- Google Compute Engine vs Scaleway