boat vs Kernel

Kernel is browser infrastructure: it hands your agent a Chromium browser on a small unikernel and deletes it when the session ends. boat hands your agent a whole Linux VM with Docker, a disk that persists, a 60fps desktop with Chromium and coding agents built in.

Where boat stands out

A whole machine, not only a browser

A boat sandbox is a full Linux VM with root, systemd, Docker and /dev/kvm, 4 vCPU / 8 GB by default and up to 16 vCPU / 32 GB. A Kernel browser is a single-purpose unikernel built to run Chromium, with 1 to 16 GB of memory and no published vCPU count. You can run commands in it, but Kernel's docs suggest uploading a binary for custom software.

Files that stay, and a fork of a running sandbox

boat keeps the disk, snapshots it every minute and on stop, and a stopped sandbox costs nothing. You can fork a running sandbox into a copy to try two paths at once. A Kernel browser is deleted after its timeout, and only cookies, storage and browser state carry over in a profile.

Coding agents and a key proxy built in

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. boat's credentials proxy keeps real API keys outside the sandbox for model providers and any HTTPS API you register. Kernel's vaults fill logins into the page, and its docs say page scripts can read them after that.

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.

Kernel

Kernel is browser infrastructure for web agents and automations: you create cloud Chromium browsers from its API and drive them over CDP.

Each Kernel browser is Chromium on its own Unikraft-based unikernel, headless (1 GB) or headful (8 or 16 GB) with Live View and a computer controls API. Around it Kernel sells stealth mode with proxies and a CAPTCHA solver, saved browser profiles, managed logins, session replays and an app runtime for your agent code. It is a different kind of product from boat: a managed browser, not a general sandbox, so this page compares where the two overlap, which is agents that use a browser.

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 rankKernel: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Kernel is a compromise here: its browser VM keeps no disk and documents no Docker, so installs, builds and tests belong on a machine like boat's.
3rd cheapest of 367
$634 a month
98th of 367, with limits
$873 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
An assistant that works all month needs files that stay; a Kernel browser's files go when it is deleted, and only cookies and storage live on in a profile.
5th cheapest of 366
$657 a month
203rd of 366, with limits
$2,033 a month
Code interpreter bursts
100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month
186th of 345, with limits
$2,000 a month
266th of 345, with limits
$347 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
300 sandboxes at once is more than Kernel's Start-Up limit of 150 browsers, and each user gets a browser rather than a machine.
Cheapest of 361
$209 a month
228th of 361, with limits
$768 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
246th of 360, with limits
$3,133 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
252nd of 364, with limits
$390 a month
CI runners
40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month
CI needs Docker and a known CPU size; Kernel documents neither, so it is a compromise.
4th cheapest of 377
$86 a month
259th of 377, with limits
$222 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
18th cheapest of 368
$26 a month
144th of 368, with limits
$33 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
Both fit: Kernel has Live View and a computer controls API for its browser, and boat streams a 60fps desktop where the agent can use any Linux app, not only the browser.
6th cheapest of 369
$200 a month
38th cheapest of 369
$1,878 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
Both fit: Kernel gives a browser over CDP with stealth built in, and boat gives a Chromium on a full VM where the agent can also run code.
7th cheapest of 367
$200 a month
21st cheapest of 367
$366 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
45th cheapest of 346
$443 a month
Mass scraping & bots (undetectable)
1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month
Kernel fits mass scraping better: stealth mode, ISP and residential proxies and a CAPTCHA solver come with the plan, while boat uses datacenter IPs and solves no CAPTCHAs.
71st of 338, with limits
$1,800 a month
4th of 338, with limits
$6,000 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
61st of 360, with limits
$293 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

FeatureboatKernel
Kind of productA full Linux VM per sandbox, with a desktop and Chromium insideA managed Chromium browser per session, with an app runtime for your code
IsolationEvery sandbox is a full VM with its own kernelEvery browser runs on its own Unikraft-based unikernel; root inside has no path to other customers
Machine size4 vCPU / 8 GB by default, up to 16 vCPU / 32 GB by plan1 GB headless, 8 or 16 GB headful; vCPU count not published
Docker and nested virtualizationDocker, systemd and /dev/kvm on every machineA single-purpose unikernel; no container runtime or /dev/kvm documented
Shell accessSSH and root on every machineSSH into the browser VM, and process exec with an as_root option
Files after a sessionThe disk persists, with snapshots every minute and on stop; a stopped sandbox costs nothingThe browser is deleted after its timeout; profiles keep cookies, storage and browser state, not files
Idle and wakeStop and resume by API; resume reboots from disk (no memory snapshots)Standby after 5 seconds idle with a memory snapshot and zero usage charge; a new connection picks it up
ForkFork a running sandbox into a copyNo fork; browser pools keep fresh browsers warm
Browser and computer use60fps desktop and Chromium, computer use and browser control, embeddable streamHeadful Chromium with Live View, a computer controls API, CDP and session replays
Bot detectionDatacenter IPs and Chromium; no CAPTCHA solvingStealth mode with an ISP proxy and a CAPTCHA solver; residential proxies; proxies not charged
Reaching the machineHTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4No inbound ports; tokenized CDP and Live View URLs, and SSH port forwarding
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 outVaults fill logins into forms, where page scripts can read them; managed auth keeps sign-ins in profiles
Coding agentsClaude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events APIA Claude Agent SDK app template that installs Claude Code when you deploy your app
RegionsEU only: Germany, Finland, Franceus-east, eu-west and ap-southeast; region choice from the Start-Up plan
Compliance and transparencySOC 2 is in progress. We're open about where we stand: live status and data retention policy.SOC 2
Every feature Battleships tracks (43)
FeatureboatKernel
Full VM (own kernel)YesYes
Docker insideYesNo
Nested virtualizationYesNo
RootYesYes
SystemdYesNot published
Desktop GUIYesYes
BrowserYesYes
Computer-use APIYesYes
SSHYesYes
Persistent diskYesNo
Pause/resumeYesYes
Automatic snapshotsYesYes
Fork/cloneYesNo
Live fork (no pause)YesNo
Memory forkNoNo
Custom image (Docker or snapshot)YesYes
≥24 h sessionsYesYes
Idle auto-stopNoYes
Wake on requestNoYes
HTTPS preview URLsYesNo
Raw TCP inboundYesNot published
Public IPv4NoNo
Egress allowlistNoYes
Secret proxyYesNo
Secret proxy for any APIYesNo
Preinstalled agentsYesYes
Agent harness APIYesNo
MCP serverNoYes
GPUNoYes
WindowsNoNo
Arm64NoNot published
Self-hostingNoNo
BYOCNoNo
Open sourceNoYes
Scoped API keysYesNot published
Spend limitsYesYes
Usage per sandbox (API)YesNo
Audit logsNoYes
SSONoYes
SOC 2NoYes
HIPAANoYes
EU data residencyYesYes
Zero data retentionYesYes

Performance

boat26.69 runs/s
KernelNot measured

There is no comparable measurement. Kernel was not in StarSling's public CPU benchmark and does not publish how many vCPUs a browser gets, so there is no number to set against boat's 26.69 runs per second at 4 vCPU / 8 GB. Kernel does publish cold starts under 20 ms for its unikernels.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Kernel 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 monthKernel, 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$5 · Free / Developer · with limits
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$26 · $20/mo$34 · Free / Developer · 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$625 · Start-Up · 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$420 · Start-Up · 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$747 · Start-Up · 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.

Kernel: Kernel bills $0.0000166667 per GB-second of memory while a browser is active: a 1 GB headless browser and an 8 GB headful one cost different amounts for the same hour. Standby and proxies are free, and plans of $0, $30 and $200 a month include $5, $10 and $50 of usage.

Limits in those scenarios

The Kernel prices in the table are for a 1 GB headless browser, the closest thing it sells, so none of them is a 4 vCPU / 8 GB machine with a 50 GB disk, and none keeps files after the browser is deleted. Kernel also caps browsers at 150 at once on Start-Up, counting standby and pool browsers, so 300 sandboxes for your users and 1,000 short jobs at once need Enterprise. On boat, the 1,000-job burst is the one that does not fit: bursts above 90 starts a minute are queued, and it needs more starts per day than the top plan allows.

ScenarioboatKernel
One coding agent, workdaysNone that bite
  • missing persistent disk
  • machine size not published
  • browser-session product, not a general sandbox
  • different kind of product: browser session
One agent that never stopsNone that bite
  • missing persistent disk
  • machine size not published
  • browser-session product, not a general sandbox
  • different kind of product: browser session
50 agents in parallelNone that bite
  • missing persistent disk, Docker inside
  • machine size not published
  • browser-session product, not a general sandbox
  • different kind of product: browser session
Sandboxes for your usersNone that bite
  • Start-Up: max 150 concurrent
  • machine size not published
  • browser-session product, not a general sandbox
  • different kind of product: browser session
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)
  • Start-Up: max 150 concurrent
  • machine size not published
  • browser-session product, not a general sandbox
  • different kind of product: browser session

Which one should you use?

Use Kernel when

your agent only needs a browser: stealth scraping behind CAPTCHAs, saved logins, session replays, or browsers in the US or Asia that cost nothing while idle.

  • Stealth, proxies and CAPTCHAs. Useful for scraping sites that block bots: Kernel's stealth mode routes through ISP proxies and solves CAPTCHAs, and proxies cost nothing extra.
  • Free standby with memory kept. Useful if your agents sit idle between steps: a Kernel browser stops charging after 5 seconds idle and comes back with its pages loaded.
  • US and Asia regions, plus browser tooling. Useful if your users are in the US or Asia, or you want saved logins and session replays without building them.

Use boat when

your agent browses and also works: installs, builds, Docker and dev servers on a full VM with a disk that stays, a 60fps desktop and coding agents built in.

  • A whole machine, not only a browser. An agent that browses and then has to install packages, run a dev server or build in Docker needs a second machine next to a Kernel browser, and you wire the two together yourself.
  • Files that stay, and a fork of a running sandbox. Downloads, repos and notes an agent made inside a Kernel browser are gone when it is deleted, so long tasks must copy files out through the file API before the timeout.
  • Coding agents and a key proxy built in. With vault fill, a prompt-injected page or an extension can read a password once it is typed in. With boat's proxy the real key never enters the machine, so there is nothing to read.

Questions people ask

Is boat cheaper than Kernel?

The two price different things: the table sets a 4 vCPU / 8 GB boat VM against a 1 GB headless Kernel browser that keeps no disk. One agent running all month comes to $26 on boat and $34 on Kernel, 50 agents in parallel to $317 against $625, and sandboxes for your users to $200 against $420. Kernel's free plan covers the workday scenario but without a disk, and for 1,000 short jobs at once neither fits.

Can a Kernel browser run Docker or a coding agent?

Kernel's browser is a single-purpose unikernel for Chromium, and its docs document no container runtime. You can run commands in it and deploy agent code as a Kernel app, including a Claude Agent SDK template. On boat each sandbox is a full VM with Docker and coding agents already installed.

Does boat have stealth mode or CAPTCHA solving like Kernel?

No. boat runs Chromium on datacenter IPs and does not solve CAPTCHAs. If you scrape sites that block bots at volume, Kernel's ISP and residential proxies and CAPTCHA solver are built for that.

Can I use boat and Kernel together?

Yes. An agent on a boat sandbox can open a Kernel browser over its API and CDP, and keep its code, files and tools on the boat VM. That puts stealth browsing on Kernel and the rest of the work on a machine that keeps its disk.

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. Kernel runs in us-east by default and offers eu-west and ap-southeast on the Start-Up and Enterprise plans.

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. Kernel lists SOC 2 and HIPAA compliance on its Enterprise plan.

Start a sandbox

Install the CLI, then run boat new.

7-day free trial

More comparisons