boat vs Azure Container Apps Sandboxes

Both give each agent its own isolated Linux machine with its own kernel. boat gives it a whole computer with SSH, nested virtualization, a desktop and coding agents built in. Azure is strongest at stateful jobs that suspend with their memory and wake on traffic, from your own image, inside an Azure account.

Where boat stands out

A whole computer, not just a place to run code

A boat sandbox is a full VM you can SSH into, with root, systemd, Docker and /dev/kvm for Android emulators and nested VMs. It has a public IPv6 address with raw ports, and HTTPS URLs on up to 50 ports. Azure's sandboxes give you a shell through the aca CLI, publish HTTPS ports only and do not document nested virtualization.

A screen and a browser for every agent

Every boat sandbox has a Linux desktop streamed at 1920x1080, 60 fps, with Chrome and computer tools for agents. You can watch an agent work and take over the mouse yourself. There is no GPU, so graphics run on the CPU. Azure's public images include no desktop or browser; you can build them into your own OCI image.

Coding agents built in

Claude Code, Codex and five other agents come preinstalled, and you prompt, steer and follow them from outside through one API. An optional paid credentials proxy keeps the real keys of supported AI providers outside the sandbox, and can switch a credential to another account when one runs out of credits.

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.

Azure Container Apps Sandboxes

Azure Container Apps Sandboxes is a service inside Microsoft Azure that gives each agent or user an isolated microVM you create, suspend and resume through an SDK.

It is part of Azure Container Apps, the hyperscaler's serverless container platform, and lives in your Azure subscription as a sandbox group resource. Each sandbox boots from an OCI image or a snapshot and has its own kernel. A stop keeps the disk, or the disk and the memory, and a memory-mode sandbox resumes in under a second. It is built for teams already on Azure that want stateful code execution with Azure networking, identity and billing.

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. “5th cheapest of 365” means 4 of the other 364 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 rankAzure Container Apps Sandboxes: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
A software factory runs installs, Docker builds, browsers and emulators for hours. boat gives each agent SSH, /dev/kvm and a desktop. On Azure you pick an image with Docker and build a desktop into your own image; nested virtualization is not documented.
5th cheapest of 365
$634 a month
33rd cheapest of 365
$5,276 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
An assistant that lives on its machine all month fits both. On boat, turn auto-stop off (a paid-plan option; the default is 1 hour), and it also gets a desktop, a browser and SSH for you to step in.
6th cheapest of 361
$657 a month
52nd cheapest of 361
$6,780 a month
Code interpreter bursts
100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month
Azure fits this well: short code runs that suspend with their memory, resume in under a second and wake again on traffic. boat counts every create as a start, so large bursts of short runs pass its plan's start limits and cost more.
189th of 336, with limits
$2,000 a month
39th cheapest of 336
$354 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
Both fit. Azure suspends idle user sandboxes with their memory. boat gives each user a full VM with up to 50 HTTPS URLs, and each create or resume counts against the plan's start limits.
3rd cheapest of 349
$209 a month
32nd cheapest of 349
$2,260 a month
RL rollouts
1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month
39th of 349, with limits
$2,000 a month
158th of 349, with limits
$10,740 a month
Agent evals
250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month
10th cheapest of 359
$200 a month
85th cheapest of 359
$1,728 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 nested virtualization. boat has /dev/kvm on every sandbox; Azure does not document it.
2nd cheapest of 374
$86 a month
108th cheapest of 374
$1,046 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A dev box wants SSH and editors that connect over SSH. boat has SSH on every sandbox. Azure gives a shell through its aca CLI, SDK and portal, and does not document SSH.
19th cheapest of 371
$26 a month
327th of 371, with limits
$262 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
Computer-use agents need a screen. boat streams a Linux desktop at 60 fps with computer tools for agents. Azure ships no desktop image, so you build and run one yourself.
7th cheapest of 367
$200 a month
58th cheapest of 367
$1,573 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
boat ships Chrome on a desktop in every sandbox. On Azure you set it up: build Chromium or Playwright into your own OCI image and import it as a disk image.
7th cheapest of 356
$200 a month
129th of 356, with limits
$795 a month
Scraping & bots (startup scale)
50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month
6th cheapest of 343
$90 a month
112th cheapest of 343
$561 a month
Mass scraping & bots (undetectable)
1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month
73rd of 327, with limits
$1,800 a month
240th of 327, with limits
$10,956 a month
Parallel red teaming
200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month
3rd cheapest of 351
$200 a month
30th cheapest of 351
$517 a month
GPU eval jobs
10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month
Not ranked: misses a requirementNot ranked: misses a requirement
macOS agents
5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month
Not ranked: misses a requirementNot ranked: misses a requirement

Key features

FeatureboatAzure Container Apps Sandboxes
IsolationFull VM with its own kernel, on physical hosts dedicated to boat (AMD Ryzen 7 9700X and Ryzen 9 9950X); vCPUs are sharedMicroVM per sandbox with its own kernel
Nested virtualizationYes, /dev/kvm for Android emulators and VMsNot documented
Docker and systemdDocker with BuildKit and Compose, systemd and root on every sandboxDepends on the image: azure-dev ships Docker on systemd and the copilot image runs systemd; the default ubuntu image has only Buildah
Custom imageStart from boat's Ubuntu 24.04 image or a saved snapshot; you cannot boot your own Docker or OCI imageAny OCI image from a public or private registry, imported as a disk image
Getting a shellSSH, plus CLI and API execInteractive shell through the aca CLI, SDK or portal; SSH is not documented
Desktop, browser and computer useLinux desktop streamed at 1920x1080, 60 fps, with Chrome and computer tools preinstalled; no GPU, graphics run on the CPUNo desktop or browser in the public images; you can build one into your own OCI image and publish it on an HTTPS port
Inbound accessHTTPS URLs for up to 50 ports, token-gated by default; SSH; a public IPv6 address with raw ports (some sandboxes have no IPv4 of their own)Public HTTPS routes per published port; private endpoint from your VNet at extra cost; no public IP
Suspend and resumeDisk is saved every minute and at each stop; memory and running programs are not kept, so resume is like a rebootStop keeps disk only, or disk and memory; a memory-mode resume takes under a second with processes running
Idle handling and wakeAuto-stop timer (default 1 hour, up to 30 days, or off on paid plans); it does not watch for idle; a stopped sandbox is started by APIAuto-suspend after an idle timeout (default 5 minutes); network traffic or a start command brings it back
ForkFork a running sandbox from its last 1-minute disk snapshotSnapshot a running sandbox with its memory and create new sandboxes from it
Outbound rulesOpen internet, no provider-side allowlistEgress policy with allow, deny, rewrite and header rules by host, path and method; VNet integration
Coding agentsSeven agents preinstalled, with a prompt, steer and events APICopilot CLI and Claude Code images, run through the exec API
Keeping AI keys out of the sandboxOptional paid credentials proxy ($1 a day while on, plus $1 per 1,000 requests): real keys of supported AI providers stay outside, and a credential can switch to another account when credits run outYou set it up: an egress Transform rule adds a group secret or managed-identity token as a header, outside the sandbox, one rule per API host; the built-in Copilot and Claude connections put tokens inside
Scaling up a fleetUp to 2,000 sandboxes at once on the $2,000 plan, self-serve; starts per minute are capped by plan (90 included and 270 at most on $2,000)300 concurrent active cores per region and subscription by default; more by Azure support request, which can need review
RegionsGermany, Finland and France; you cannot pick a region; no US or Asia region40 Azure regions across the Americas, Europe, Asia, Africa and Oceania
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)
FeatureboatAzure Container Apps Sandboxes
Full VM (own kernel)YesYes
Docker insideYesYes
Nested virtualizationYesNot published
RootYesYes
SystemdYesYes
Desktop GUIYesWith your own setup
BrowserYesWith your own setup
Computer-use APIYesNo documented way
SSHYesNo documented way
Persistent diskYesYes
Pause/resumeYesYes
Automatic snapshotsYesYes
Fork/cloneYesYes
Live fork (no pause)YesYes
Memory forkNo documented wayYes
Custom image (Docker or snapshot)YesYes
≥24 h sessionsYesYes
Idle auto-stopNo documented wayYes
Wake on requestNo documented wayYes
HTTPS preview URLsYesYes
Raw TCP inboundYesNo documented way
Public IPv4No documented wayNo documented way
Egress allowlistNo documented wayYes
Secret proxyYesYes
Secret proxy for any APIYesYes
Preinstalled agentsYesYes
Agent harness APIYesNo documented way
MCP serverNo documented wayNo documented way
GPUNo documented wayNo documented way
WindowsNo documented wayNo documented way
Arm64No documented wayNo documented way
Self-hostingNo documented wayNo documented way
BYOCNo documented wayNo documented way
Open sourceNo documented wayNo documented way
Scoped API keysYesNot published
Spend limitsYesYes
Usage per sandbox (API)YesYes
Audit logsNo documented wayYes
SSONo documented wayYes
SOC 2No documented wayYes
HIPAANo documented wayYes
EU data residencyNo documented wayYes
Zero data retentionYesNot published

Performance

boat28.06 runs/s
Azure Container Apps SandboxesNot measured

Azure Container Apps Sandboxes is not in the public benchmarks we use, so we cannot say which is faster. For reference, boat did 28.06 runs per second on the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, second of 15 in StarSling's run of 3 October 2026 (run 37073680030). Every boat sandbox in this run ran on an AMD Ryzen 9 9950X host. boat also places some sandboxes on AMD Ryzen 7 9700X hosts, which this run did not measure.

V8 Web Tooling Benchmark at a 4 vCPU / 8 GB target, runs per second, higher is better. Source: StarSling's run of 3 October 2026 (run 37073680030), median of 5 sandboxes. Providers ran on different hardware in different regions. StarSling gives a shared rank when the run cannot separate two providers. Every boat sandbox in this run ran on an AMD Ryzen 9 9950X host. boat also places some sandboxes on AMD Ryzen 7 9700X hosts, which this run did not measure. Azure Container Apps Sandboxes is not in the public benchmarks we use, so we cannot say which is faster.

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. Start limits, per-start fees and plan minimums are in the price, boat's included.

Scenarioboat, a monthAzure Container Apps Sandboxes, 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$68 · Pay-as-you-go
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$26 · $20/mo$262 · Pay-as-you-go · 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$3,421 · Pay-as-you-go
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$989 · Pay-as-you-go
1,000 short jobs at once
2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month
$8,800 · $2000/mo · with limits$2,148 · Pay-as-you-go

boat: Plans cost $20, $100, $500 or $2,000 a month, and the whole plan price comes back as VM time for that month. A card is required, also for the 7-day trial. A default sandbox (4 vCPU / 8 GB) costs $0.036 an hour, billed per second while it runs, busy or idle. The $20 plan includes 200 starts a day (12 a minute); an extra start costs $0.03 to $0.07, and starts above 6 times the included number (3 times per minute) fail. A stopped sandbox has no compute charge and keeps its disk for at least 14 days. Saved snapshots above the included count ($0.056 a day each) and the optional AI proxy still charge while sandboxes are stopped.

Azure Container Apps Sandboxes: Azure bills per second on the Container Apps Consumption meters for the vCPU and memory of the sandbox's tier while it runs, idle time before auto-suspend included. The XL tier (4 cores / 8 GiB) is $0.432 an hour at the East US rates. Stopped sandboxes pay no CPU or memory, and there is no plan fee. Storage for images, snapshots and volumes will be billed per GiB (announced as coming soon). Azure does not state whether its savings plans cover Sandboxes.

Limits in those scenarios

One agent that never stops: Azure bills a running sandbox every second, so its lowest price there ($262 a month) needs a 3-year savings plan, and Azure does not state that the plan covers Sandboxes. boat runs it for $26 on the $20 plan once auto-stop is off (default 1 hour). 50 agents in parallel: Docker on Azure means the azure-dev image or your own image; the default ubuntu image has only Buildah. 50 XL sandboxes use 200 of Azure's default 300 active cores per region and subscription. Sandboxes for your users: 300 L sandboxes at once use 600 cores, so Azure needs a support request first, and the increase can need review. 1,000 short jobs at once: boat's $2,000 plan charges for starts above 90 a minute and refuses them above 270, and boat never queues a refused start, so boat does not fit that one and it would cost about $8,800. Azure costs $2,148 there, but 1,000 L sandboxes use 2,000 active cores, so it also needs a quota raise first.

ScenarioboatAzure Container Apps Sandboxes
One coding agent, workdaysNone for this scenarioNone for this scenario
One agent that never stopsNone for this scenario
  • 3-year commitment
50 agents in parallelNone for this scenarioNone for this scenario
Sandboxes for your usersNone for this scenarioNone for this scenario
1,000 short jobs at once
  • 2,400 starts/day included, paid above that (needs ~6,667)
  • 600 starts/hour included, paid above that (needs ~1,000)
  • starts above 90/min are paid, and starts above 270/min fail with 429
None for this scenario
boat's own limits
  • Plans start at $20 a month. The plan price comes back as VM time for that month. Unused time does not carry over. Source
  • Starts (create, fork, resume): the $20 plan includes 12 a minute, 60 an hour and 200 a day. The $2,000 plan includes 90 a minute, 600 an hour and 2,400 a day. Extra starts cost $0.03 to $0.07. A start fails with 429 above 3 times the included starts per minute, or 6 times per hour or per day. Source
  • A stopped sandbox has no compute charge. boat keeps its disk for at least 14 days. A saved snapshot keeps it longer; extra saved snapshots and the AI credentials proxy are paid. Source
  • Auto-stop defaults to 1 hour. You can turn it off on a paid plan. The trial allows 2 hours at most. Source
  • boat saves the disk every minute and at each stop. Memory, running programs and /tmp are not saved: a resume is like a reboot. A machine failure loses writes after the last good snapshot. Source
  • Machines run in Germany, Finland and France only. Snapshot copies can be stored outside the EU. Source
  • Public IPv6 and outbound IPv4. No dedicated IPv4. The IP changes at each resume. Source
  • Hosts are AMD Ryzen 7 9700X and Ryzen 9 9950X. vCPUs are shared. No GPU. You cannot boot your own Docker image. Source

Which one should you use?

Use Azure Container Apps Sandboxes when

your team already runs on Azure and wants sandboxes from its own OCI images that suspend with their memory, wake on traffic, sit behind strict egress rules, or run outside the EU.

  • Memory suspend and wake on traffic. Good for stateful jobs that sleep between requests: a sandbox suspends with its memory and resumes in under a second.
  • Fine-grained outbound rules. Its egress policy can allow, deny or rewrite each outbound HTTP request by host, path and method, with VNet integration.
  • Your own image, inside Azure, in 40 regions. Teams already on Azure get Entra ID, Azure billing, VNets and their own OCI images, in 40 regions on six continents.

Use boat when

your agents need a whole computer for real work: SSH, Docker, emulators, a desktop and a browser, with coding agents built in, at a lower price in four of the five scenarios on this page.

  • A whole computer, not just a place to run code. On Azure, nested virtualization is not documented, so an Android emulator or a VM is not a supported path. Tools that expect SSH, like remote editors, go through the aca CLI or the exec API instead.
  • A screen and a browser for every agent. On Azure, a computer-use or browser agent needs you to build a browser, a virtual display and a streaming stack into an image, then keep them working. Azure publishes no example for this.
  • Coding agents built in. On Azure, the Copilot and Claude images run through the generic exec API, and their built-in connections put tokens inside the sandbox. To keep a key outside, you write an egress Transform rule for each API host yourself. boat's proxy, once on, does this for supported model providers and can switch accounts when credits run out.

Questions people ask

Is boat cheaper than Azure Container Apps Sandboxes?

In four of the five scenarios on this page, boat costs less: for example $20 against $68 a month for one agent on workdays, and $317 against $3,421 for 50 agents in parallel. For 1,000 short jobs at once, boat's start limits make it cost more ($8,800 against $2,148) and refuse starts above 270 a minute, so Azure fits that one better, after a quota raise. A sandbox that sits idle most of the time can also cost less on Azure, because Azure suspends it after 5 idle minutes, while a running boat sandbox bills at full rate until it stops.

Can I run Docker inside an Azure Container Apps sandbox?

Yes, with the right image. Microsoft's azure-dev image lists Docker on a systemd base, the ubuntu image has Buildah for building Dockerfiles, and you can import your own image. boat runs Docker with BuildKit and Compose on every sandbox and also exposes /dev/kvm for Android emulators and VMs, which Azure does not document.

Can I SSH into an Azure sandbox?

Azure does not document SSH. It gives you an interactive shell through the aca CLI, the SDK or the portal, and published ports are HTTPS routes. Every boat sandbox has an SSH server and an SSH endpoint, so editors and tools that connect over SSH work as they do on any server.

Can boat suspend a sandbox with its memory like Azure?

No. boat saves the disk every minute and once more at each stop, so files survive, but memory and running programs are not kept and a resume is like a reboot. If a machine fails, writes after the last good snapshot are lost. Azure can suspend with memory and resume running processes in under a second.

Can I use my own container image?

On Azure, yes: any OCI image from a public or private registry can be imported as a disk image. On boat, no: you start from boat's Ubuntu 24.04 image or from a saved snapshot, and install what you need.

Do I need an Azure account to use it?

Yes. Azure Container Apps Sandboxes needs an Azure subscription, a Microsoft Entra ID account and the Container Apps SandboxGroup Data Owner role. boat signs you in with GitHub, Google or an emailed code, and asks for a card, also for the trial.

Is boat SOC 2 compliant?

SOC 2 is in progress, so boat has no SOC 2 report today. No HIPAA BAA and no SAML SSO are documented. Measured sandbox uptime is 99.95% over all sandboxes, and no SLA is published. Live status is at status.ascii.dev and the data retention policy is at docs.boat.dev/data-retention.

Start a sandbox

Install the CLI, then run boat new.

7-day free trial

More comparisons