boat vs Bitrise

Bitrise is a mobile CI/CD service: it runs iOS and Android builds on Apple silicon Macs and Linux VMs, and now rents remote dev environments for coding agents. boat is a sandbox for agents: a full Linux VM that runs with no time cap, keeps its files with snapshots every minute, and lets you drive the agents inside from outside.

Where boat stands out

A machine that keeps running

A boat sandbox has no session cap on paid plans, so an agent can work for days. Its disk is snapshotted every minute and on stop, and a stopped sandbox costs nothing. Bitrise CI builds time out at 210 minutes and roll back, and its dev sessions keep one disk that is not backed up.

Agents you drive from outside

boat ships Claude Code, Codex, OpenCode and other agents in each sandbox, and you send prompts, steer them and read their events through an API. Bitrise starts Claude Code from its CLI and you work with it in a terminal. boat's credentials proxy also keeps real AI keys out of the VM.

Fork, Docker and open ports

Every boat sandbox has root, systemd, Docker and /dev/kvm, and you can fork a running one into a copy. It can serve HTTPS on up to 50 ports and take raw TCP. Bitrise dev sessions are reached by SSH or VNC, have no public URL for a port, and macOS sessions cannot run Docker.

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.

Bitrise

Bitrise is a mobile CI/CD platform that builds, tests and ships iOS and Android apps on its own Mac and Linux machines.

Bitrise is a different kind of product: a CI service. Each build runs in its own virtual machine from a Bitrise stack, and the machine is rolled back after the build. Since 2026 it also sells Remote Dev Environments, macOS and Linux sessions with a persistent disk where you can start Claude Code from a CLI or drive the machine through an MCP server. It is built for mobile 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. “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 rankBitrise: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Bitrise CI builds stop at a timeout and roll back, so a software factory there means its remote dev environments, where each tier runs only 1 to 5 machines at once.
3rd cheapest of 367
$634 a month
38th of 367, with limits
$3,403 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
An always-on assistant needs a machine that never times out; Bitrise sessions stop after days of inactivity, and boat runs with no session cap on paid plans.
5th cheapest of 366
$657 a month
73rd of 366, with limits
$1,970 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
289th of 345, with limits
$658 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
Bitrise dev environment tiers run at most 5 VMs at once and CI Pro 10 macOS or 30 Linux builds, far below the hundreds of sandboxes this needs.
Cheapest of 361
$209 a month
269th of 361, with limits
$1,604 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
271st of 360, with limits
$6,818 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
289th of 364, with limits
$1,274 a month
CI runners
40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month
Bitrise is a real CI service, best for iOS and Android pipelines; boat fits generic Linux jobs an agent kicks off, with Docker and no build quota.
4th cheapest of 377
$86 a month
181st of 377, with limits
$859 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
Both keep a persistent disk; Bitrise's monthly cap makes a Mac dev box predictable, while boat's Linux box keeps snapshots every minute.
18th cheapest of 368
$26 a month
168th of 368, with limits
$119 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
Bitrise lets an agent take screenshots of and control a macOS desktop through its MCP server; boat streams a 60fps Linux desktop with Chromium.
6th cheapest of 369
$200 a month
68th of 369, with limits
$1,186 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
7th cheapest of 367
$200 a month
132nd of 367, with limits
$746 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
131st of 346, with limits
$862 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
256th of 338, with limits
$13,088 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
73rd of 360, with limits
$537 a month
GPU eval jobs
10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month
Neither has GPUs.
Can't run itCan't run it
macOS agents
5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month
Only Bitrise fits: it rents Apple silicon Macs with iOS simulators, and boat has no macOS.
Can't run it21st of 36, with limits
$400 a month

Key features

FeatureboatBitrise
Kind of productSandbox for AI agents: a Linux VM you create, run, stop and fork by APIMobile CI/CD service, plus Remote Dev Environments for developers and agents
IsolationEvery sandbox is a full VM with its own kernel, on dedicated AMD Ryzen 9 9950X (Zen 5) hostsEvery build or session is its own VM; CI VMs are rolled back after each build
Operating systemsLinux only; Wine for Windows appsmacOS on Apple silicon with iOS simulators, and Linux with Android emulators
Docker and nested virtualizationDocker, root, systemd and /dev/kvm on every machineDocker and nested virtualization on Linux machines, so Android emulators run there; macOS sessions have neither, because Apple's virtualization has no nested virtualization
How long a machine runsNo session cap on paid plansCI builds time out at 210 minutes (up to 4 hours on Enterprise); dev sessions have no time cap but auto-terminate after 5 days idle by default (up to 30 days or off), and maintenance can stop them now and then
Files after a stopSnapshots every minute and on stop; a stopped sandbox costs nothingDev sessions keep one persistent disk, which is not backed up; CI machines keep nothing
ForkFork a running sandbox into an independent copyNo fork; warm pools hand out fresh sessions booted from a template
Coding agentsClaude Code, Codex, OpenCode and more preinstalled, driven from outside with a prompt, steer and events APIClaude Code started from the CLI and used interactively; other agents through warmup scripts; an event list for agent stops and permission prompts
Desktop and computer use60fps Linux desktop and Chromium, computer use and browser control, embeddable streamVNC desktop on macOS sessions only; an open-source MCP server takes screenshots and controls it
Keeping keys out of the sandboxAI credentials proxy for model providers and any HTTPS API; real keys never enter the VM; switch to another account when credits run outSecrets are environment variables, so the value enters the machine
Reaching the machineHTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4SSH, VNC and a web terminal; device preview links stream a simulator; no public URL for a port
Mobile build pipelineNone built in; you install the tools you needWorkflows, code signing, build cache, release management and over-the-air updates
RegionsEU only: Germany, Finland, FranceUS, Ireland, Frankfurt, Singapore and Sydney
Self-hostingManaged onlyBitrise on AWS in your own account, or fully self-hosted Bitrise Runners
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)
FeatureboatBitrise
Full VM (own kernel)YesYes
Docker insideYesYes
Nested virtualizationYesYes
RootYesYes
SystemdYesNo
Desktop GUIYesYes
BrowserYesYes
Computer-use APIYesYes
SSHYesYes
Persistent diskYesYes
Pause/resumeYesYes
Automatic snapshotsYesNo
Fork/cloneYesNo
Live fork (no pause)YesNo
Memory forkNoNo
Custom image (Docker or snapshot)YesYes
≥24 h sessionsYesNo
Idle auto-stopNoYes
Wake on requestNoNo
HTTPS preview URLsYesNo
Raw TCP inboundYesNo
Public IPv4NoNo
Egress allowlistNoNo
Secret proxyYesNo
Secret proxy for any APIYesNo
Preinstalled agentsYesYes
Agent harness APIYesYes
MCP serverNoYes
GPUNoNo
WindowsNoNo
Arm64NoYes
Self-hostingNoYes
BYOCNoYes
Open sourceNoNo
Scoped API keysYesNot published
Spend limitsYesNot published
Usage per sandbox (API)YesNo
Audit logsNoNot published
SSONoYes
SOC 2NoYes
HIPAANoNo
EU data residencyYesYes
Zero data retentionYesNot published

Performance

boat26.69 runs/s
BitriseNot measured

There is no comparable measurement: Bitrise has no result in the public sandbox benchmark we use, so we do not compare speeds. boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB in StarSling's run of 2026-09-29.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Bitrise 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 monthBitrise, 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$260 · Pro · with limits
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$26 · $20/mo$119 · RDE Pro · 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$2,309 · Pro · 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$812 · Pro · 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$1,538 · Pro · 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.

Bitrise: Bitrise CI is sold by plan: Starter is a flat fee with unlimited minutes on Medium machines, and Pro adds per-minute machine time and a monthly build allowance on top of its fee. Remote Dev Environments charge a base fee with some included minutes, then per minute up to a monthly cap per OS, after which that OS is free for the rest of the month.

Limits in those scenarios

Bitrise limits machines by plan: Pro runs at most 10 macOS or 30 Linux builds at once, and a dev environment tier runs 1 to 5 VMs at once, so 50 agents in parallel and 300 user sandboxes do not fit. CI builds stop at the build timeout, so a job that runs all day has to be split up. Dev environments have no such timeout; they stop after a stretch of inactivity you can set, and now and then for maintenance. On boat the limit is a sandbox count (100 to 2,000 by plan) plus starts per minute, so the 1,000-job burst gets queued and needs more starts per day than the $2,000 plan allows.

ScenarioboatBitrise
One coding agent, workdaysNone that bite
  • Pro: sessions capped at 3.5 h
  • Pro: Linux M size not allowed
  • different kind of product: CI runner
  • egress price not published
One agent that never stopsNone that bite
  • different kind of product: dev-environment
  • egress price not published
50 agents in parallelNone that bite
  • Pro: max 10 concurrent
  • Pro: sessions capped at 3.5 h
  • different kind of product: CI runner
  • egress price not published
Sandboxes for your usersNone that bite
  • Pro: max 10 concurrent
  • different kind of product: CI runner
  • 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)
  • Pro: max 10 concurrent
  • different kind of product: CI runner
  • egress price not published

Which one should you use?

Use Bitrise when

you build and ship iOS or Android apps, need macOS on Apple silicon, or must run in the US, Asia Pacific or your own AWS account.

  • Macs and iOS simulators. Useful if your agents build or test iOS apps: Bitrise rents Apple silicon Macs with iOS simulators and a desktop.
  • A full mobile CI/CD pipeline. Useful if you ship mobile apps: Bitrise handles builds, code signing, caching, store releases and over-the-air updates.
  • More regions and self-hosting. Useful if you need machines in the US or Asia Pacific, or inside your own AWS account.

Use boat when

your agents need a full Linux VM that runs for days with Docker, snapshots every minute, forks, open ports and coding agents you drive from outside.

  • A machine that keeps running. An agent that runs past a build timeout loses the job and the machine's state. On Bitrise you move long work into a dev session, and a lost disk there has no restore point.
  • Agents you drive from outside. Without a prompt and events API, running many agents means scripting a terminal session for each one. On Bitrise, secrets are environment variables, so a key sits inside a machine the agent controls.
  • Fork, Docker and open ports. An agent that builds a web app needs to show it running at a URL. On Bitrise you would set up your own tunnel, and you cannot copy a session to try two ideas side by side.

Questions people ask

Is boat cheaper than Bitrise?

In the scenarios we priced, boat costs less in four of five: one coding agent on workdays comes to $20 a month on boat and $260 on Bitrise, one agent running all month to $26 versus $119, and 50 agents in parallel to $317 versus $2,309. For 1,000 short jobs at once, boat comes to $2,000 and Bitrise to $1,538, and neither fits without higher limits.

Can I run AI coding agents on Bitrise?

Yes, in Remote Dev Environments. bitrise-cli rde claude creates a session, clones your branch and starts Claude Code, and other agents can be installed with a warmup script. On boat the agents come preinstalled, and you drive them through an API instead of a terminal.

Does boat run macOS or iOS simulators?

No. boat runs Linux VMs only, so iOS builds and simulators need a Mac service like Bitrise. boat does run Docker and /dev/kvm, so Linux tools and Android emulators work inside a sandbox.

Is Bitrise a sandbox for agents?

Not at its core. Bitrise is a mobile CI/CD service whose build machines are reset after each build, and its Remote Dev Environments are cloud machines for developers and agents with a fixed number of VMs per tier. boat is built to create, stop and fork many agent sandboxes by API.

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. Bitrise runs in the US, Ireland, Frankfurt, Singapore and Sydney, and can also run in your own AWS account.

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. Bitrise lists SOC 2 Type II and PCI DSS on its security page.

Start a sandbox

Install the CLI, then run boat new.

7-day free trial

More comparisons