Azure Container Apps Sandboxes vs smol machines

Both give each agent its own microVM with its own kernel and can copy a running machine with its memory. Azure Container Apps Sandboxes lives inside your Azure account with fine-grained outbound rules and 40 regions. smol machines is an open-source engine that runs the same VM on your laptop, in its hosted cloud or on your own servers, with a managed agents API.

The companies and products

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.

smol machines

smol machines makes isolated Linux microVMs that run the same way on your laptop, in their hosted cloud, or on your own servers.

The core is smolvm, an open-source microVM engine built on libkrun. Each machine has its own guest kernel. smol cloud is the paid hosted service. It runs machines behind a REST API, Node, Python and Rust SDKs, and a managed agents API. It is for developers who want one VM model from local development to production, with live branching of running machines.

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. “33rd cheapest of 365” means 32 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 workloadAzure Container Apps Sandboxes: price ranksmol machines: price rankboat: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
33rd cheapest of 365
$5,276 a month
105th of 365, with limits
$604 a month
5th cheapest of 365
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
52nd cheapest of 361
$6,780 a month
41st cheapest of 361
$1,038 a month
6th cheapest of 361
$657 a month
Code interpreter bursts
100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month
Both fit short code runs. smol machines has a one-minute minimum and bills the CPU and memory the guest uses. Azure suspends with memory and resumes in under a second.
39th cheapest of 336
$354 a month
16th cheapest of 336
$140 a month
189th of 336, with limits
$2,000 a month
Sandboxes for your users
300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month
Both fit sandboxes for your users, and both bring a stopped machine back on the next request. boat adds a desktop and up to 50 HTTPS URLs, and each start counts against its plan's start limits.
32nd cheapest of 349
$2,260 a month
12th cheapest of 349
$531 a month
3rd cheapest of 349
$209 a month
RL rollouts
1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month
Both can copy a running machine with its memory, so many rollouts start from one prepared state. smol machines branches pay the one-minute minimum once rather than per clone. A boat fork boots fresh from the last 1-minute disk snapshot.
158th of 349, with limits
$10,740 a month
119th of 349, with limits
$6,978 a month
39th of 349, with limits
$2,000 a month
Agent evals
250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month
Both fit agent evals. smol machines has a registry image with the Harbor eval harness, and Azure has Copilot CLI and Claude Code images run through exec.
85th cheapest of 359
$1,728 a month
49th cheapest of 359
$670 a month
10th cheapest of 359
$200 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 boots VMs needs nested virtualization. smol cloud exposes it on its Startup ($150 a month plus usage) and Enterprise plans, Azure does not document it, and boat has /dev/kvm on every plan.
108th cheapest of 374
$1,046 a month
69th cheapest of 374
$514 a month
2nd cheapest of 374
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A dev box wants SSH; neither documents an SSH route, and smol's WebSocket terminal is capped at one hour per connection. Azure also bills every running second until it suspends. boat has SSH on every sandbox.
327th of 371, with limits
$262 a month
230th of 371, with limits
$29 a month
19th cheapest of 371
$26 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. On both you build a desktop and a stream into your own image. boat streams a Linux desktop at 60 fps with computer tools.
58th cheapest of 367
$1,573 a month
135th of 367, with limits
$156 a month
7th cheapest of 367
$200 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
Neither ships a managed browser: on Azure you build one into your image, and smol machines has a guide for running your own Chromium. boat ships Chrome in every sandbox.
129th of 356, with limits
$795 a month
58th of 356, with limits
$136 a month
7th cheapest of 356
$200 a month
Scraping & bots (startup scale)
50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month
112th cheapest of 343
$561 a month
97th cheapest of 343
$242 a month
6th cheapest of 343
$90 a month
Mass scraping & bots (undetectable)
1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month
240th of 327, with limits
$10,956 a month
145th of 327, with limits
$4,344 a month
73rd of 327, with limits
$1,800 a month
Parallel red teaming
200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month
30th cheapest of 351
$517 a month
59th of 351, with limits
$377 a month
3rd cheapest of 351
$200 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 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 requirementNot ranked: misses a requirement

Key features

FeatureAzure Container Apps Sandboxessmol machinesboat
IsolationMicroVM per sandbox with its own kernellibkrun microVM with its own guest kernelFull VM with its own kernel, on hosts dedicated to boat (AMD Ryzen 7 9700X and Ryzen 9 9950X); vCPUs are shared
Biggest machineSeven tiers up to 16 cores / 32 GiB / 320 GiB diskUp to 32 vCPU / 32 GB on Startup, 16 vCPU / 24 GB on ProUp to 16 vCPU / 32 GB on the $100 plan and above
Stop, suspend and wakeSuspend with memory and disk, resume in under a second with processes running; traffic wakes itFilesystem and image survive a stop, /tmp is emptied; checkpoints save a running machine with its RAM; a command or request starts it again; stopped disk bills at $0.0001/GB-hourDisk saved every minute and at stop; no compute charge while stopped; disk kept at least 14 days; a resume is like a reboot
Copy a running machineSnapshot a running sandbox with its memory and create new sandboxes from itBranch a running machine with its memory and processes; checkpoints capture RAM to durable storageFork a running sandbox from its last 1-minute disk snapshot; the fork boots fresh
Nested virtualizationNot documentedYes on Startup ($150 a month plus usage) and Enterprise, with nestedVirt; refused on Standard and Pro/dev/kvm on every plan for Android emulators and VMs
Docker insideDepends on the image: azure-dev ships Docker on systemd; the default ubuntu image has only BuildahDocker runs inside the VM, with its data on the machine's ext4 diskDocker, systemd and root on every sandbox
Getting a shellInteractive shell through the aca CLI, SDK or portal; SSH is not documentedCommand API, sessions and a WebSocket terminal capped at one hour per connection; no SSH route documentedSSH, plus a command API and terminal
Inbound accessPublic HTTPS routes per published port; private endpoint from your VNet; no public IPUp to 16 published ports via token URLs or an authenticated connect route, plus a TCP tunnel route; no dedicated IPUp to 50 HTTPS URLs, token-gated by default; public IPv6 with raw ports; the IP changes at each resume
Outbound rulesAllow, deny, rewrite and header rules by host, path and method; VNet integrationBlocked, open, or an allowlist of hosts and CIDR rangesOpen internet, no provider-side allowlist
Desktop, browser and computer useNo desktop or browser in the public images; you can build one into your own OCI imageYou set it up: run Chromium from an OCI image (official guide); no hosted desktop or computer-use route1920x1080, 60 fps desktop with Chrome and computer tools
Coding agentsCopilot CLI and Claude Code images, run through the exec APIManaged agents API: Claude Code, Codex, OpenCode or any command as the harness, with turns, cancel, event stream, pause/resume, rewind to a turn and branchSeven harnesses installed, with a prompt, steer and events API
Keeping API keys out of the sandboxYou set it up: an egress Transform rule adds a group secret as a header from outside, one rule per API host; the built-in Copilot and Claude connections put tokens insideStored credentials bound to a machine: the guest sees a placeholder, and the real value is swapped in on HTTPS to allowed hosts; plain env secrets are readableOptional AI credentials proxy ($1 a day while on, plus $1 per 1,000 requests): real keys of supported providers stay outside; other secrets enter the sandbox
Regions40 Azure regions, 9 of them in the EUNo region choice in the cloud APIGermany, Finland and France; you cannot pick a region; no US or Asia region
Self-hostingNo; runs only inside an Azure subscriptionOpen-source smolvm runs on your laptop or your own servers with the same .smolmachine artifactNo; hosted service only
Every feature Battleships tracks (43)
FeatureAzure Container Apps Sandboxessmol machinesboat
Full VM (own kernel)YesYesYes
Docker insideYesYesYes
Nested virtualizationNot publishedYesYes
RootYesNot publishedYes
SystemdYesNot publishedYes
Desktop GUIWith your own setupNo documented wayYes
BrowserWith your own setupWith your own setupYes
Computer-use APINo documented wayNo documented wayYes
SSHNo documented wayNo documented wayYes
Persistent diskYesYesYes
Pause/resumeYesYesYes
Automatic snapshotsYesNo documented wayYes
Fork/cloneYesYesYes
Live fork (no pause)YesYesYes
Memory forkYesYesNo documented way
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesYesYes
Idle auto-stopYesYesNo documented way
Wake on requestYesYesNo documented way
HTTPS preview URLsYesYesYes
Raw TCP inboundNo documented wayNot publishedYes
Public IPv4No documented wayNo documented wayNo documented way
Egress allowlistYesYesNo documented way
Secret proxyYesYesYes
Secret proxy for any APIYesYesYes
Preinstalled agentsYesYesYes
Agent harness APINo documented wayYesYes
MCP serverNo documented wayYesNo documented way
GPUNo documented wayNo documented wayNo documented way
WindowsNo documented wayNo documented wayNo documented way
Arm64No documented wayYesNo documented way
Self-hostingNo documented wayYesNo documented way
BYOCNo documented wayYesNo documented way
Open sourceNo documented wayYesNo documented way
Scoped API keysNot publishedNot publishedYes
Spend limitsYesYesYes
Usage per sandbox (API)YesYesYes
Audit logsYesNot publishedNo documented way
SSOYesYesNo documented way
SOC 2YesNo documented wayNo documented way
HIPAAYesNo documented wayNo documented way
EU data residencyYesNo documented wayNo documented way
Zero data retentionNot publishedNot publishedYes

Performance

Azure Container Apps SandboxesNot measured
smol machinesNot measured
boat28.06 runs/s

Azure Container Apps Sandboxes and smol cloud are not in the public benchmarks we use, so we cannot say which is faster. In StarSling's run of 3 October 2026 (run 37073680030), boat did 28.06 runs/s on the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, second of 15. 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. smol machines publishes a boot time under 200 ms for smolvm, which is a start time, not CPU speed.

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 and smol machines are 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.

ScenarioAzure Container Apps Sandboxes, a monthsmol machines, a monthboat, 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
$68 · Pay-as-you-go$25 · Standard$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$262 · Pay-as-you-go · with limits$93 · Standard$26 · $20/mo
50 agents in parallel
50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside
$3,421 · Pay-as-you-go$1,139 · Pro$317 · $20/mo × 3 seats
Sandboxes for your users
2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once
$989 · Pay-as-you-go$465 · Startup$200 · $100/mo × 2 seats
1,000 short jobs at once
2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month
$2,148 · Pay-as-you-go$1,389 · Startup$8,800 · $2000/mo · with limits

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, including idle time before auto-suspend (5 minutes by default). 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.

smol machines: smol cloud charges $0.04 for every hour a machine runs plus $0.05 per active vCPU-hour and $0.0162 per active GB-hour of memory, with a one-minute minimum. Disk bills at $0.0001 per GB-hour, also while stopped, and egress at $0.05 per GB. Pro ($20 a month) and Startup ($150 a month) are fees on top of usage, not credit.

smol machines: Plans cost $20, $100, $500 or $2,000 a month, and the whole plan price comes back as VM time for that month; unused time does not carry over. 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 12 starts a minute, 60 an hour and 200 a day; extra starts cost $0.03 to $0.07 each. A stopped sandbox has no compute charge and keeps its disk for at least 14 days. Saved snapshots above the included count and the optional AI credentials proxy still charge while stopped.

Limits in those scenarios

Azure has no session cap, but it bills a running sandbox every second, so one agent that never stops reaches its lowest price only with a 3-year savings plan, and Azure does not state that the plan covers Sandboxes. Azure's default quota is 300 active cores per region and subscription: 50 XL sandboxes use 200, but 300 user sandboxes and 1,000 short jobs need a support request first. Docker inside means picking an image like azure-dev. smol machines caps machines by plan, 20 on Standard, 100 on Pro and 1,000 on Startup, with stopped machines still holding a slot, so 50 agents needs Pro and 1,000 short jobs at once needs Startup. boat does not fit the 1,000-job burst: the $2,000 plan includes 90 starts a minute, starts above 270 a minute fail with 429, and boat never queues a start.

ScenarioAzure Container Apps Sandboxessmol machinesboat
One coding agent, workdaysNone for this scenarioNone for this scenarioNone for this scenario
One agent that never stops
  • 3-year commitment
None for this scenarioNone for this scenario
50 agents in parallelNone for this scenarioNone for this scenarioNone for this scenario
Sandboxes for your usersNone for this scenarioNone for this scenarioNone for this scenario
1,000 short jobs at onceNone for this scenarioNone for this scenario
  • 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

Which one should you use?

Use Azure Container Apps Sandboxes when

your team already runs on Azure and wants sandboxes that suspend with their memory, wake on traffic, sit behind per-request egress rules, or run in a region outside the EU.

  • Memory suspend and wake on traffic. For stateful jobs that sleep between requests: a sandbox suspends with its memory and resumes in under a second with processes running.
  • Fine-grained outbound rules. Its egress policy can allow, deny or rewrite each outbound request by host, path and method, and join your Azure VNets.
  • Inside Azure, in 40 regions. Teams already on Azure get Entra ID sign-in, Azure billing and their own networks, in 40 regions around the world.

Use smol machines when

you want one open-source VM engine from your laptop to your own servers, live branches of a running machine, a managed agents API, or 32 vCPU machines.

  • Open source, laptop to your own servers. One open-source engine, smolvm, runs the same machine on your laptop, in smol cloud or on servers you run yourself.
  • Live branches and a managed agents API. Branch a running machine with its memory and processes, paying the one-minute minimum once, and drive agents with turns, rewind and branch.
  • Lower bills and bigger machines. smol machines costs less than Azure in all five scenarios we priced, and Startup machines go up to 32 vCPU / 32 GB.

Where boat fits

Azure Container Apps Sandboxes and smol machines both do well at short, stateful microVMs that sleep, wake and copy their memory, which boat does not do. If your agents instead work for hours like a person at a computer, installing tools, running emulators, using a browser and serving apps, boat gives each one a whole machine with that built in, within its plan start limits.

SSH and /dev/kvm on every plan

Neither Azure nor smol cloud documents SSH, so SSH tools must be rewired to an exec API. Azure does not document nested virtualization, and smol cloud has it only on Startup and Enterprise. boat has SSH and /dev/kvm on every sandbox.

A screen and a browser for every agent

On both you build, package and maintain a browser, a virtual display and a stream in your own image. boat streams a Linux desktop at 60 fps with Chrome and computer tools ready.

A disk saved every minute

Azure and smol cloud save state when you snapshot or checkpoint. boat saves each sandbox's disk every minute and at each stop, with no setup, and keeps a stopped disk for at least 14 days.

Use boat when your agents need a whole computer for hours of work: SSH, Docker, /dev/kvm on every plan, a desktop and a browser, and coding agents you drive from outside, within its plan start limits. Compare it one to one: boat vs Azure Container Apps Sandboxes and boat vs smol machines.

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

Questions people ask

Is Azure Container Apps Sandboxes cheaper than smol machines?

No, smol machines costs less in all five scenarios on this page, for example $25 against $68 a month for one coding agent on workdays. Azure bills the full tier while a sandbox runs, while smol bills the CPU and memory the guest uses plus a base fee per running hour. boat comes to $20 for the same scenario.

Which is faster, Azure Container Apps Sandboxes or smol machines?

Neither is in the public benchmarks we use, so we cannot say. boat did 28.06 runs/s on the V8 Web Tooling Benchmark at 4 vCPU / 8 GB in StarSling's run of 3 October 2026, on an AMD Ryzen 9 9950X host.

Can I run Docker and nested VMs in Azure Container Apps Sandboxes and smol machines?

Docker, yes on both, with care: Azure's azure-dev image ships Docker on systemd, while the default ubuntu image has only Buildah; smol says Docker data must live on the machine's ext4 disk. Nested virtualization is on smol cloud's Startup and Enterprise plans and not documented on Azure. boat has Docker and /dev/kvm on every plan.

Can I SSH into either one?

Neither documents SSH. Azure gives a shell through the aca CLI, SDK or portal, and smol cloud gives a command API and a WebSocket terminal capped at one hour per connection. Every boat sandbox takes SSH.

Can I self-host smol machines or Azure Container Apps Sandboxes?

smol machines, yes: the smolvm engine is open source and runs on macOS, Linux and Windows hosts. Azure Container Apps Sandboxes runs only inside an Azure subscription, and boat is a hosted service only.

Do they keep API keys out of the sandbox?

Yes, in different ways. smol cloud binds stored credentials so the guest sees only a placeholder and the node swaps in the real value on HTTPS to allowed hosts. Azure can add a secret as a header from outside with an egress rule you write per host, but its built-in Copilot and Claude connections put tokens inside. boat's optional paid proxy keeps real keys of supported AI providers outside.

Try boat

Install the CLI, then run boat new.

7-day free trial

More comparisons