boat vs Ubicloud

A Ubicloud VM is a plain Linux machine that you set up, secure and keep running yourself, and it keeps billing while stopped. boat also gives each agent a full VM, but with snapshots every minute, stops that cost nothing, fork of a running machine, a desktop and coding agents built in.

Where boat stands out

Snapshots every minute, free stops, fork while running

boat saves a sandbox's disk every minute and when it stops, and a stopped sandbox costs nothing. You can fork a running sandbox in one call. On Ubicloud, a stopped VM keeps billing as if it were running, and copying a VM means stopping it and capturing a machine image of a single disk up to 40 GiB.

Coding agents and a credentials proxy built in

Every boat sandbox has Claude Code, Codex, OpenCode and other agents ready, and you drive them from outside with a prompt, steer and events API. The credentials proxy keeps real AI keys out of the machine and can switch to another account when one runs out of credits.

A desktop and HTTPS ports with no setup

boat streams a 60fps desktop with Chromium for computer use, and each sandbox can serve up to 50 ports on token-protected HTTPS URLs, plus raw TCP. A Ubicloud VM has a public IP, and HTTPS means running TLS on the VM yourself, even behind a load balancer, which only supplies the certificate.

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.

Ubicloud

Ubicloud is an open source cloud that runs VMs, managed Postgres, Kubernetes and GitHub Actions runners on leased bare-metal servers.

You rent Linux VMs (Standard with dedicated CPU, or cheaper Burstable with shared CPU) from Ubicloud's managed service, or run the same AGPL code on your own servers. Around the VMs it offers private subnets, an inbound firewall, load balancers, managed PostgreSQL, Kubernetes, GPU VMs, inference endpoints and cheap GitHub Actions runners. It is general infrastructure, not an agent sandbox, so this page compares the two where they overlap: giving an agent its own Linux machine.

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 rankUbicloud: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Both give a full Linux VM with root and Docker, but a Ubicloud VM starts as an empty Ubuntu, Debian or AlmaLinux image and bills while stopped, while boat comes with agents, snapshots and free stops.
3rd cheapest of 367
$634 a month
126th of 367, with limits
$1,136 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
A Ubicloud VM can run an assistant around the clock, but you build the agent setup, the HTTPS link and the backups yourself; boat has them built in.
5th cheapest of 366
$657 a month
154th of 366, with limits
$1,139 a month
Code interpreter bursts
100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month
Ubicloud fits one-minute runs better: its cheap burstable VMs bill per minute with no plan fee, while boat's start limits make big bursts a compromise.
186th of 345, with limits
$2,000 a month
59th of 345, with limits
$52 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
Ubicloud's default quota is 256 vCPU once a project is verified, so 300 user sandboxes of 2 vCPU need a raise from support; boat allows 100 to 2,000 sandboxes by plan.
Cheapest of 361
$209 a month
53rd of 361, with limits
$390 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
234th of 360, with limits
$1,560 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
31st cheapest of 364
$624 a month
CI runners
40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month
Ubicloud sells managed GitHub Actions runners, x64 and arm64, billed per minute of job time; boat runs CI jobs on general sandboxes with Docker and /dev/kvm.
4th cheapest of 377
$86 a month
28th cheapest of 377
$300 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
Both fit a 24/7 dev box; a Ubicloud Standard VM gives dedicated CPU, while a boat machine keeps its disk with snapshots every minute and costs less here.
18th cheapest of 368
$26 a month
67th cheapest of 368
$91 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
boat streams a 60fps desktop with Chromium and computer use; on Ubicloud you install the desktop, the browser and the control loop on a plain VM yourself.
6th cheapest of 369
$200 a month
132nd of 369, with limits
$312 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
188th of 367, with limits
$156 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
165th of 346, with limits
$78 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
60th of 338, with limits
$1,573 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
129th of 360, with limits
$81 a month
GPU eval jobs
10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month
Only Ubicloud fits: it rents Nvidia GPU VMs on request, and boat has no GPUs.
Can't run it93rd of 98, with limits
$795 a month
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

FeatureboatUbicloud
What you getA sandbox API for agents, on full VMsA general cloud VM you set up and run yourself, plus Postgres, Kubernetes and CI runners
IsolationEvery sandbox is a full VM with its own kernelEvery VM is a full KVM guest (Cloud Hypervisor) with its own kernel
CPUDedicated AMD Ryzen 9 9950X (Zen 5) hosts; 26.69 runs/s at 4 vCPU / 8 GBStandard VMs have dedicated CPU; Burstable VMs get 50% of each vCPU and burst above it; no comparable measurement
Docker and nested virtualizationDocker, root, systemd and /dev/kvm on every machine, documentedRoot over SSH, so Docker installs like on any Linux VM; nested virtualization is not documented
StoppingA stopped sandbox costs nothing and keeps its diskA stopped VM keeps its resources reserved and billing continues as if it were running
SnapshotsAutomatic snapshots every minute and on stopMachine images of a stopped VM's disk, made by hand; no automatic snapshots
ForkFork a running sandbox in one call (disk, no memory)Stop the VM, capture a machine image (single disk up to 40 GiB), then boot new VMs from it
Reaching the machineHTTPS URLs for up to 50 ports, token-protected; raw TCP; own IPFree public IPv6, public IPv4 for $3 a month; for HTTPS the VM runs TLS itself; a Layer 4 load balancer can supply a ZeroSSL certificate but does not terminate TLS
Desktop and computer use60fps desktop and Chromium, computer use and browser control, embeddable streamNone; you install a desktop and browser on the VM yourself
Coding agentsClaude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events APINone; you install and drive agents yourself, for example over SSH
Keeping keys out of the machineAI credentials proxy for model providers and any HTTPS API you register; switch a credential to another account when credits run outNo credentials proxy; secrets live on the VM
Network rulesOpen internet on every plan; no provider-side allowlistStateful inbound firewall by IP range and port; outbound rules are not supported
GPUs, arm64 and self-hostingNo GPUs, x86 only, managed service onlyNvidia GPU VMs on request, arm64 CI runners, and the whole stack is AGPL code you can run yourself
RegionsEU only: Germany, Finland, FranceGermany, Finland and Virginia (US), plus GPUs in Istanbul
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)
FeatureboatUbicloud
Full VM (own kernel)YesYes
Docker insideYesYes
Nested virtualizationYesNot published
RootYesYes
SystemdYesNot published
Desktop GUIYesNo
BrowserYesNo
Computer-use APIYesNo
SSHYesYes
Persistent diskYesYes
Pause/resumeYesNo
Automatic snapshotsYesNo
Fork/cloneYesYes
Live fork (no pause)YesNo
Memory forkNoNo
Custom image (Docker or snapshot)YesYes
≥24 h sessionsYesYes
Idle auto-stopNoNo
Wake on requestNoNo
HTTPS preview URLsYesNo
Raw TCP inboundYesNot published
Public IPv4NoYes
Egress allowlistNoNo
Secret proxyYesNo
Secret proxy for any APIYesNo
Preinstalled agentsYesNo
Agent harness APIYesNo
MCP serverNoNo
GPUNoYes
WindowsNoNo
Arm64NoYes
Self-hostingNoYes
BYOCNoYes
Open sourceNoYes
Scoped API keysYesNot published
Spend limitsYesNot published
Usage per sandbox (API)YesNo
Audit logsNoYes
SSONoYes
SOC 2NoYes
HIPAANoNo
EU data residencyYesYes
Zero data retentionYesNot published

Performance

boat26.69 runs/s
UbicloudNot measured

There is no comparable measurement for Ubicloud: its VMs were not part of the public StarSling benchmark run where boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB. Results would also depend on the family you pick, since Standard VMs have dedicated CPU and Burstable VMs get half of each vCPU with short bursts.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Ubicloud 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 monthUbicloud, 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$22 · Usage
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$26 · $20/mo$91 · Usage
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$1,098 · Usage
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$156 · Usage
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$312 · Usage · 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.

Ubicloud: Ubicloud bills each VM per minute for as long as it exists, running or stopped, and invoices monthly; a public IPv4 adds $3 a month. So the workday price assumes you delete the VM each night and lose its disk. The cheapest shapes are Burstable VMs with shared CPU.

Limits in those scenarios

Ubicloud limits the total vCPU a project can run: 32 for a new project and 256 once it has paid an invoice over $5, with more from support. 50 agents of 4 vCPU fit in 256, but 300 user sandboxes of 2 vCPU and 1,000 short jobs at once do not without a raise. The cheap burstable shape used for those also guarantees only half of each vCPU, so a busy burst gets throttled. 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.

ScenarioboatUbicloud
One coding agent, workdaysNone that biteNone that bite
One agent that never stopsNone that biteNone that bite
50 agents in parallelNone that biteNone that bite
Sandboxes for your usersNone that biteNone that bite
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)
  • burstable CPU: 50% of each vCPU guaranteed, throttled above that

Which one should you use?

Use Ubicloud when

you want an open source cloud you can also run yourself, GPU VMs, cheap GitHub Actions runners, a US region, or a long-lived VM next to managed Postgres and Kubernetes.

  • Open source you can run yourself. Useful if you want to run the cloud on your own servers: Ubicloud's whole stack is AGPL code you can deploy on leased bare metal.
  • GPUs on request. Useful if your agents train or serve models: Ubicloud rents Nvidia A100, B200 and RTX PRO 6000 VMs once support enables them.
  • A whole cloud, with cheap CI and a US region. Useful if the agent's VM must sit next to managed Postgres, Kubernetes and GitHub Actions runners, or run in the US.

Use boat when

you want each agent on its own full VM that is ready to work: coding agents, a desktop, HTTPS ports and minute-by-minute snapshots built in, fork while running, and stops that cost nothing.

  • Snapshots every minute, free stops, fork while running. Agents sit idle between tasks. On Ubicloud you either pay full price for the idle VM or delete it and lose its disk, and copying a busy agent's machine means stopping its work first.
  • Coding agents and a credentials proxy built in. A Ubicloud VM starts as an empty OS image, so you install, update and supervise each agent yourself, and the agent can read any key you put on the VM.
  • A desktop and HTTPS ports with no setup. An agent that tests a web app needs a browser it can see and a link it can share. On Ubicloud you build the desktop, the browser and the TLS setup for every VM yourself.

Questions people ask

Is boat cheaper than Ubicloud?

For agents that run long, yes: one agent that never stops comes to $26 a month on boat against $91 on Ubicloud, and 50 agents in parallel come to $317 against $1,098. One coding agent on workdays is about even, at $20 on boat and $22 on Ubicloud if you delete the VM every night. For 300 user sandboxes Ubicloud's shared-CPU burstable VMs come to $156 against $200 on boat.

Do I pay for a Ubicloud VM that is stopped?

Yes. Ubicloud's changelog says a stopped VM keeps its configuration, storage and compute reserved, and billing continues as if it were still running. To stop paying you delete the VM. On boat, a stopped sandbox costs nothing and keeps its disk.

Is Ubicloud open source?

Yes. Ubicloud's code is on GitHub under the AGPL v3, and you can run it yourself on leased bare-metal servers or use its managed service. boat is a managed service only.

Does Ubicloud have an agent sandbox product?

No. Ubicloud sells general VMs, managed Postgres, Kubernetes, CI runners, GPU VMs and inference endpoints, with an API, a CLI and a web console. You install agents yourself and drive them over SSH. boat is built for agents, with coding agents, a desktop and a credentials proxy in every sandbox.

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. Ubicloud runs VMs in Germany, Finland and Virginia in the US, and GPU VMs in Istanbul 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. Ubicloud says in its security docs that it is SOC 2 Type 2 compliant.

Start a sandbox

Install the CLI, then run boat new.

7-day free trial

More comparisons