boat vs OpenPond

OpenPond is an open-source agent workspace with cloud sandboxes you drive through its TypeScript SDK, up to 10 at a time below Enterprise. boat gives every agent a full Linux VM with SSH, root, Docker, /dev/kvm, a desktop and coding agents built in, with no 10-sandbox ceiling.

Where boat stands out

A whole Linux machine, not just an SDK

On boat each agent gets a VM you can SSH into, with root, systemd, Docker and /dev/kvm, plus raw TCP and its own IP. OpenPond sandboxes are also microVMs, but you reach them through its SDK, CLI or MCP: exec, processes, PTY sessions and HTTPS preview ports. SSH, systemd and /dev/kvm are not documented there.

Fleets bigger than ten

OpenPond's Free plan runs 1 cloud sandbox at a time, and Starter and Pro run up to 10. Anything more needs an Enterprise contract. On boat, 50 agents in parallel and 300 sandboxes for your users fit on self-serve plans.

Coding agents and a desktop built in

boat ships Claude Code, Codex, OpenCode and other agents in each sandbox, and you drive them from outside with a prompt, steer and events API. Each VM also streams a 60fps desktop with Chromium for computer use. OpenPond's Work runs its own hosted model, and its cloud sandboxes are headless.

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.

OpenPond

OpenPond is an open-source workspace where people and AI agents do tasks, with paid cloud sandboxes on the side.

The main product is a desktop, web, CLI and chat workspace with an agent harness, evals and a training workbench, released under the MIT license. Its cloud sandboxes are Firecracker microVMs that your server code creates and controls through the openpond-sdk TypeScript SDK: run commands, manage files, open preview ports, snapshot and fork. It is built for teams that want either a hosted model to finish a task (Work) or raw sandboxes their own code drives.

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 workloadboat: price rankOpenPond: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
A software factory runs many agents with Docker and services at once; OpenPond stops at 10 sandboxes below Enterprise, and boat adds /dev/kvm and SSH on each machine.
3rd cheapest of 367
$634 a month
292nd of 367, with limits
$5,787 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
Always-on assistants need many machines that never stop; OpenPond fits only up to 10 of them on self-serve plans.
5th cheapest of 366
$657 a month
327th of 366, with limits
$8,768 a month
Code interpreter bursts
100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month
OpenPond fits short code runs better: it bills only running seconds with no plan fee for the size, while boat's start limits make big bursts a compromise.
186th of 345, with limits
$2,000 a month
148th of 345, with limits
$391 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
300 sandboxes for your users need far more than OpenPond's 10 at once, so they need an Enterprise contract there.
Cheapest of 361
$209 a month
163rd of 361, with limits
$2,471 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
156th of 360, with limits
$11,874 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
186th of 364, with limits
$1,918 a month
CI runners
40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month
OpenPond has no runner integration or SSH, so CI jobs run through its SDK; boat machines take a runner like any Linux server.
4th cheapest of 377
$86 a month
192nd of 377, with limits
$1,152 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
18th cheapest of 368
$26 a month
339th of 368, with limits
$351 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
OpenPond's cloud sandboxes are headless; its browser control drives the browser in its local desktop app, while boat streams a 60fps desktop with Chromium from each VM.
6th cheapest of 369
$200 a month
288th of 369, with limits
$1,727 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
Browser agents need a browser inside the sandbox; OpenPond documents none in its cloud sandboxes, and boat ships Chromium with browser control.
7th cheapest of 367
$200 a month
297th of 367, with limits
$868 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
287th of 346, with limits
$582 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
243rd of 338, with limits
$11,642 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
287th of 360, with limits
$579 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 it
macOS agents
5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month
Can't run itCan't run it

Key features

FeatureboatOpenPond
IsolationEvery sandbox is a full VM with its own kernelFirecracker microVM with its own kernel
Root, systemd and nested virtualizationRoot, systemd, Docker and /dev/kvm on every machineDocker commands are recognized; root, systemd and /dev/kvm are not documented
How you get inSSH and SCP, plus an API and a CLISDK, CLI and MCP: exec, processes and PTY sessions; no SSH
Sandboxes at onceNo 10-sandbox ceiling; large fleets on self-serve plans1 on Free, up to 10 on Starter and Pro, more only on Enterprise
Machine size4 vCPU / 8 GB by default, 8 vCPU from the $20 planFive published sizes from 0.25 vCPU / 0.5 GiB to 4 vCPU / 8 GiB; templates allow up to 8 vCPU / 32 GiB
Files after a stopSnapshots every minute and on stop; a stopped sandbox costs nothingStopping ends compute billing and keeps the filesystem; snapshots are made by hand
Snapshot and forkFork a running sandbox; named snapshots as templates (disk, not memory)Filesystem snapshots you can fork, restore or publish as templates (disk, not memory)
Custom imagesNamed snapshots (templates), shareable by code; no Dockerfile buildsStart from a base, a snapshot, an OCI image or a Dockerfile
Reaching the sandboxHTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4HTTPS preview ports, private with tokens or public; no raw TCP or public IP documented
Outbound internetOpen internet on every plan; no provider-side allowlistEgress set to restricted, allow or block, with allowed hosts, in the sandbox network policy
Desktop and browser60fps desktop and Chromium in each VM, with computer use and browser controlCloud sandboxes are headless; browser tools drive the local desktop app
Coding agentsClaude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events APIWork runs an OpenPond-hosted model with file and command tools; third-party coding agents are not preinstalled
Keeping keys out of the sandboxAI credentials proxy for model providers and any HTTPS API you register; switch a credential to another account when credits run outOpenPond Connect links Google, GitHub and X accounts without handing raw tokens to agents
Spend capsAccount spend limits on top of monthly plan creditPer-sandbox budget, spend, duration and idle limits set in code
Compliance and transparencySOC 2 is in progress. We're open about where we stand: live status and data retention policy.Not published
Every feature Battleships tracks (43)
FeatureboatOpenPond
Full VM (own kernel)YesYes
Docker insideYesYes
Nested virtualizationYesNo
RootYesNot published
SystemdYesNot published
Desktop GUIYesNo
BrowserYesNo
Computer-use APIYesNo
SSHYesNo
Persistent diskYesYes
Pause/resumeYesYes
Automatic snapshotsYesNo
Fork/cloneYesYes
Live fork (no pause)YesYes
Memory forkNoNo
Custom image (Docker or snapshot)YesYes
≥24 h sessionsYesYes
Idle auto-stopNoYes
Wake on requestNoNo
HTTPS preview URLsYesYes
Raw TCP inboundYesNot published
Public IPv4NoNo
Egress allowlistNoYes
Secret proxyYesNo
Secret proxy for any APIYesNo
Preinstalled agentsYesNo
Agent harness APIYesNo
MCP serverNoYes
GPUNoNo
WindowsNoNo
Arm64NoNo
Self-hostingNoNo
BYOCNoNo
Open sourceNoYes
Scoped API keysYesYes
Spend limitsYesYes
Usage per sandbox (API)YesNo
Audit logsNoNot published
SSONoYes
SOC 2NoNot published
HIPAANoNot published
EU data residencyYesNot published
Zero data retentionYesNot published

Performance

boat26.69 runs/s
OpenPondNot measured

There is no comparable measurement for OpenPond: it was not part of the public CPU benchmark boat was measured in. boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB on dedicated AMD Ryzen 9 9950X (Zen 5) hosts in StarSling's run.

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

Scenarioboat, a monthOpenPond, 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$76 · Free
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$26 · $20/mo$351 · Free
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$3,799 · Starter · with limits
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$1,083 · Starter · with limits
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$2,375 · Starter · with limits

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.

OpenPond: OpenPond bills per second for each vCPU, GiB of memory and GiB of VM disk while a sandbox runs, and retained volumes and snapshots keep costing storage after it stops. Starter and Pro include $10 or $20 of monthly usage, and beyond that you add credits or turn on overage.

Limits in those scenarios

The limit that bites on OpenPond is sandboxes at once: 1 on Free and 10 on Starter and Pro. That is why 50 agents in parallel, 300 sandboxes for your users and 1,000 short jobs at once do not fit without Enterprise. OpenPond does not publish an egress price, so heavy downloads are an open cost. On boat the 1,000-job burst is limited by starts: bursts above 90 starts a minute are queued, and it needs more starts per day than the $2,000 plan allows.

ScenarioboatOpenPond
One coding agent, workdaysNone that bite
  • egress price not published
One agent that never stopsNone that bite
  • egress price not published
50 agents in parallelNone that bite
  • Starter: max 10 concurrent
  • egress price not published
Sandboxes for your usersNone that bite
  • Starter: max 10 concurrent
  • egress price not published
1,000 short jobs at once
  • 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)
  • Starter: max 10 concurrent
  • egress price not published

Which one should you use?

Use OpenPond when

you want an open-source agent workspace with a hosted Work API, or a handful of sandboxes that boot from Dockerfiles with strict egress rules and per-sandbox spend caps.

  • Egress rules and spend caps per sandbox. Useful if untrusted code must stay offline or inside a host list, with a hard dollar cap on each sandbox.
  • Build from a Dockerfile or OCI image. Useful if your environments already live as Dockerfiles or registry images: OpenPond boots a sandbox straight from them.
  • An open-source workspace around the sandbox. Useful if you want a MIT-licensed agent workspace, harness and evals that you can read, change and run locally.

Use boat when

your agents need real Linux machines in numbers: SSH, root, Docker and /dev/kvm, a desktop and browser, coding agents you drive by API, and more than 10 running at once.

  • A whole Linux machine, not just an SDK. Without SSH and root, tools that expect a normal server, like remote editors, CI runners or emulators that need /dev/kvm, have to be wrapped in SDK calls on OpenPond, or they do not run at all.
  • Fleets bigger than ten. An agent team, an eval run or a product that gives each user a sandbox hits OpenPond's 10-sandbox wall fast, and the extra jobs wait or need a sales call.
  • Coding agents and a desktop built in. If you want Claude Code or Codex on OpenPond, you install it and write the code that starts, prompts and watches it yourself. Browser and desktop agents have no screen to use in its cloud sandboxes.

Questions people ask

Is boat cheaper than OpenPond?

In the scenarios we priced at 4 vCPU / 8 GB, yes: one coding agent on workdays comes to $20 a month on boat and $76 on OpenPond, and one agent running all month comes to $26 versus $351. 50 agents in parallel come to $317 versus $3,799, and OpenPond also caps self-serve plans at 10 sandboxes. OpenPond's default 1 vCPU sandbox is small, so very light jobs can cost less there.

Can I SSH into an OpenPond sandbox?

OpenPond documents no SSH. You run commands, start processes and open PTY sessions through its SDK, CLI or MCP server. boat machines take SSH like any Linux server.

How many OpenPond sandboxes can run at once?

One on Free, and up to 10 on Starter and Pro. More needs an Enterprise contract with dedicated capacity.

Does OpenPond have a desktop or browser in the sandbox?

Not in its cloud sandboxes, which are headless Linux. Its browser tools control the browser inside the local OpenPond desktop app. boat streams a 60fps desktop with Chromium from each VM, with computer use and browser control.

Is OpenPond open source?

The workspace, CLI, SDK, harness and evals are MIT licensed on GitHub. The cloud sandboxes themselves run on OpenPond's hosted service and need an API key. boat is a hosted service too.

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