CodeSandbox SDK vs smol machines

Both give each sandbox its own microVM with its own kernel. CodeSandbox SDK is strongest at VMs that hibernate with their memory and wake on the next HTTP request, while smol machines is strongest at an open-source engine you can run anywhere, live branches of a running machine and outbound allowlists.

The companies and products

CodeSandbox SDK

CodeSandbox SDK is an API from CodeSandbox, now a Together AI company, that starts sandboxed development VMs on the same infrastructure as the CodeSandbox editor.

Each sandbox is a Firecracker microVM built from a template or any Dockerfile. When it goes idle it hibernates with a memory snapshot, and it resumes or forks from that snapshot in seconds. You drive it from a TypeScript SDK through terminals, commands, files and HTTPS hosts. Together's docs now call it the legacy offering and point new users to its successor, the together-sandbox SDK.

Self-hosting
Not offered

smol machines

smol machines makes fast, 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, where each machine has its own guest kernel. smol cloud is the paid hosted service that runs fleets of those machines behind a REST API and Node, Python and Rust SDKs. It is built 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. “46th cheapest of 345” means 45 of the other 344 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 workloadCodeSandbox SDK: price ranksmol machines: price rankboat: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Both run Docker inside the VM, but neither has SSH or /dev/kvm; boat has both, plus coding agents you prompt and steer through an API.
178th of 367, with limits
$1,939 a month
73rd of 367, with limits
$604 a month
3rd cheapest of 367
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
215th of 366, with limits
$2,478 a month
148th of 366, with limits
$1,038 a month
5th cheapest of 366
$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 well: CodeSandbox SDK wakes hibernated VMs with their memory, and smol machines has a one-minute minimum that branches of a warm machine pay only once.
46th cheapest of 345
$642 a month
12th cheapest of 345
$140 a month
186th of 345, 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
smol machines fits up to 300 user sandboxes at once on Startup, while CodeSandbox SDK's Scale plan stops at 250 VMs; boat fits too and adds a desktop and raw TCP for each user.
140th of 361, with limits
$1,707 a month
7th cheapest of 361
$531 a month
Cheapest of 361
$209 a month
RL rollouts
1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month
Both can fork a running sandbox with its memory, so many rollouts start from one prepared state; boat forks a running sandbox with its disk.
348th of 360, with limits
$3,861 a month
111th of 360, with limits
$6,978 a month
34th of 360, with limits
$2,000 a month
Agent evals
250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month
64th cheapest of 364
$1,335 a month
36th cheapest of 364
$670 a month
10th cheapest of 364
$200 a month
CI runners
40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month
Neither offers nested virtualization on its hosted VMs, so CI jobs that build Android apps or boot VMs need boat's /dev/kvm.
92nd cheapest of 377
$860 a month
66th cheapest of 377
$514 a month
4th cheapest of 377
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A box that runs all day is billed by the hour on both, and a stopped smol machine still pays for its disk, while a stopped boat sandbox costs nothing.
310th of 368, with limits
$240 a month
216th of 368, with limits
$29 a month
18th cheapest of 368
$26 a month
Computer-use agents
50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month
Neither CodeSandbox SDK nor smol machines hosts a desktop, so you build your own stack; boat streams a Linux desktop at 60fps with computer use built in.
206th of 369, with limits
$691 a month
90th of 369, with limits
$156 a month
6th cheapest of 369
$200 a month
Browser-use agents
50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month
CodeSandbox SDK has a headless Chromium template and smol machines has a guide for running your own Chromium; boat ships Chromium on a full desktop.
251st of 367, with limits
$443 a month
176th of 367, with limits
$136 a month
7th cheapest of 367
$200 a month
Scraping & bots (startup scale)
50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month
281st of 346, with limits
$508 a month
241st of 346, with limits
$242 a month
6th cheapest of 346
$90 a month
Mass scraping & bots (undetectable)
1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month
193rd of 338, with limits
$7,390 a month
148th of 338, with limits
$4,344 a month
71st of 338, with limits
$1,800 a month
Parallel red teaming
200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month
240th of 360, with limits
$326 a month
252nd of 360, with limits
$377 a month
2nd cheapest of 360
$200 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 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 itCan't run it

Key features

FeatureCodeSandbox SDKsmol machinesboat
IsolationFirecracker microVM with its own kernellibkrun microVM with its own guest kernelFull VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
Biggest machineUp to 16 cores / 32 GB on Scale, 64 cores / 128 GB on EnterpriseUp to 32 vCPU / 32 GB on Startup, 16 vCPU / 24 GB on ProUp to 16 vCPU / 32 GB on the $100 plan and above
Pause and resumeHibernates with a memory snapshot; an HTTP request wakes it with processes still running; no charge while hibernatedA stop keeps the filesystem and empties /tmp and memory; stopped disk keeps billing; the next command starts it againDisk is saved every minute and on stop; a stopped sandbox costs nothing; resume boots it again by API
ForkFork a hibernated sandbox from its snapshot, or a running one sharing its memoryBranch a running machine with its memory and processes; checkpoints capture RAM to a fileFork a running sandbox with its disk
Docker and rootDocker and root; built from any DockerfileDocker inside the machine, with its data on the ext4 diskDocker, systemd and root
Nested virtualizationNot documentedNot exposed in the cloud API; a local smolvm flag onlyYes, /dev/kvm for Android emulators and VMs
Shell accessTerminals, commands and files through the SDK; no SSHCommand API, sessions and a WebSocket terminal capped at one hour per connection; no SSHSSH, plus exec and files over the API
Inbound accessHTTPS host per port, public by default, signed tokens for private hosts; no raw TCP or public IPUp to 16 published ports through a returned URL or an authenticated connect route; no dedicated IPHTTPS URLs for up to 50 ports, token-protected by default; raw TCP; a dedicated IPv4 or IPv6
Outbound rulesNo outbound allowlistBlocked, open, or an allowlist of hosts and CIDR rangesOpen internet, no provider-side allowlist
Desktop and browserNo desktop; a headless Chromium templateNo hosted desktop; a guide shows running your own Chromium inside a machineLinux desktop streamed at 60fps, Chromium and computer use built in
Coding agentsRun your own agent through the command APIRegistry images with the Codex CLI and the Harbor eval harness, run through the command APISeven agents preinstalled, including Claude Code and Codex, with a prompt, steer and events API
Keeping API keys out of the sandboxSecrets passed in as environment variablesCloud secrets are environment variables the guest can read; placeholder credential swapping only in self-hosted smolvmCredentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out
RegionsA country hint steers scheduling, for example to Europe, but placement is not guaranteedNo region choice in the cloud APIEU only (Germany, Finland, France), with EU data residency
Self-hostingNot offeredOpen-source smolvm runs on your laptop or your own servers, with the same .smolmachine artifactNo, hosted only
Every feature Battleships tracks (43)
FeatureCodeSandbox SDKsmol machinesboat
Full VM (own kernel)YesYesYes
Docker insideYesYesYes
Nested virtualizationNoNoYes
RootYesNot publishedYes
SystemdNot publishedNot publishedYes
Desktop GUINoNoYes
BrowserNoNoYes
Computer-use APINoNoYes
SSHNoNoYes
Persistent diskYesYesYes
Pause/resumeYesYesYes
Automatic snapshotsYesNoYes
Fork/cloneYesYesYes
Live fork (no pause)YesYesYes
Memory forkYesYesNo
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesYesYes
Idle auto-stopYesYesNo
Wake on requestYesYesNo
HTTPS preview URLsYesYesYes
Raw TCP inboundNoNot publishedYes
Public IPv4NoNoNo
Egress allowlistNoYesNo
Secret proxyNoNoYes
Secret proxy for any APINoNoYes
Preinstalled agentsNoYesYes
Agent harness APINoNoYes
MCP serverNoYesNo
GPUNoNoNo
WindowsNoNoNo
Arm64NoYesNo
Self-hostingNoYesNo
BYOCNoYesNo
Open sourceNoYesNo
Scoped API keysNot publishedNot publishedYes
Spend limitsNot publishedYesYes
Usage per sandbox (API)NoYesYes
Audit logsNot publishedNot publishedNo
SSOYesYesNo
SOC 2YesNoNo
HIPAANoNoNo
EU data residencyNoNoYes
Zero data retentionNot publishedNot publishedYes

Performance

CodeSandbox SDKNot measured
smol machinesNot measured
boat26.69 runs/s

There is no comparable CPU measurement for either CodeSandbox SDK or smol machines: neither was part of the StarSling hpc-sandbox-benchmarks run. boat did 26.69 runs per second on that Node.js build-and-test loop at 4 vCPU / 8 GB, on dedicated AMD Ryzen 9 9950X hosts. smol machines publishes a boot time under 200ms for smolvm, which is a start-time figure, not CPU speed.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. CodeSandbox SDK and smol machines have 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.

ScenarioCodeSandbox SDK, 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
$193 · Scale$25 · Standard$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$341 · Scale$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
$2,500 · Scale$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
$815 · Scale · with limits$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
$1,632 · Scale · with limits$1,389 · Startup$2,000 · $2000/mo · with limits

CodeSandbox SDK: You buy VM credits and each size burns a fixed number per hour, billed per started minute while the VM runs, including the idle time before it hibernates. Scale adds a $170 monthly platform fee that is not usage credit, and a hibernated VM costs nothing.

smol machines: smol cloud charges a base fee for every hour a machine runs, plus the CPU and memory the guest actually uses, with a one-minute minimum and disk billed even while stopped. Pro ($20 a month) and Startup ($150 a month) are fees on top of usage, not credit.

smol machines: Plans start at $20 a month, and the whole fee is usage credit. You pay per second while a sandbox runs, and a stopped sandbox costs nothing and keeps its disk.

Limits in those scenarios

CodeSandbox SDK's Scale plan allows 250 VMs at once, so 300 user sandboxes and 1,000 short jobs at once do not fit without Enterprise; each VM has a 20 GB disk, and the price for more disk and for egress is not published. 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 needs Startup. boat has no such cap at those sizes, but at 1,000 short jobs at once even its $2,000 plan queues starts above 90 a minute.

ScenarioCodeSandbox SDKsmol machinesboat
One coding agent, workdays
  • disk beyond 20 GiB: price unknown
  • egress price not published
None that biteNone that bite
One agent that never stops
  • disk beyond 20 GiB: price unknown
  • egress price not published
None that biteNone that bite
50 agents in parallel
  • egress price not published
None that biteNone that bite
Sandboxes for your users
  • Scale: max 250 concurrent
  • egress price not published
None that biteNone that bite
1,000 short jobs at once
  • Scale: max 250 concurrent
  • egress price not published
None that bite
  • 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)

Which one should you use?

Use CodeSandbox SDK when

your sandboxes sleep between visits and should wake on the next HTTP request with their memory intact, and you are fine building on what Together now calls its legacy SDK.

  • Hibernation with memory, woken by a request. Good for previews and code runs used in bursts: an idle VM sleeps with its memory and wakes on the next HTTP request.
  • Nothing billed while asleep. A hibernated CodeSandbox SDK VM costs nothing, while a stopped smol machine still pays for its disk.
  • Very large machines on Enterprise. Teams that need one big machine can get up to 64 cores and 128 GB through Enterprise sales.

Use smol machines when

you want one open-source VM engine from your laptop to your own servers, fan out live branches of a running machine, or must limit outbound traffic to a list of hosts.

  • Open source, from laptop to your own servers. Teams that want one VM engine locally, in a hosted cloud and on their own servers can use the open-source smolvm.
  • Big bursts and many user sandboxes. Startup holds up to 1,000 machines, so 300 user sandboxes or 1,000 short jobs at once fit where CodeSandbox SDK's Scale stops at 250.
  • Outbound allowlists and arm64. Teams that must limit which hosts a machine reaches, or need arm64 images, get both on smol cloud.

Where boat fits

CodeSandbox SDK and smol machines are both good at small VMs that sleep, wake and fork fast. If what you are really running is agents that work for hours, with Docker, emulators, a browser and servers other people reach, boat gives each one a whole machine with the agent and its keys already set up.

A whole machine, not just a VM boundary

Neither offers /dev/kvm or SSH on its hosted VMs, so Android emulators and nested VMs cannot start, and every tool goes through an SDK or exec call. smol's terminal also drops each connection after one hour.

Desktop, browser and agents built in

Neither hosts a desktop, so a computer-use agent waits until you build, package and maintain a desktop, Chromium and a screen stream. boat streams a 60fps desktop and ships seven coding agents you drive from outside by API.

Keys that never enter the sandbox

Both hand secrets to the guest as environment variables, so one prompt injection can leak them. smol's credential swapping works only if you run smolvm on your own servers; boat's credentials proxy covers any HTTPS API on the hosted service.

Use boat when your agents do real work for hours: Docker, emulators, SSH, a desktop and a browser, servers on up to 50 ports, coding agents you drive from outside and keys that never enter the sandbox. Compare it one to one: boat vs CodeSandbox SDK and boat vs smol machines.

Questions people ask

Is CodeSandbox SDK cheaper than smol machines?

No, smol machines comes out cheaper in all five scenarios on this page. One coding agent on workdays is about $193 a month on CodeSandbox SDK against $25 a month on smol machines, and 50 agents in parallel is $2,500 a month against $1,139 a month. boat is lower still in four of the five.

Which is faster, CodeSandbox SDK or smol machines?

Neither has a comparable CPU measurement, so there is no fair speed ratio between them. boat did 26.69 runs per second on the StarSling Node.js benchmark at 4 vCPU / 8 GB. smol machines publishes a boot time under 200ms for smolvm, which measures start time, not CPU speed.

Can CodeSandbox SDK and smol machines run Docker?

Yes, both run Docker inside the VM. smol's guide says Docker data must live on the machine's ext4 disk. Neither exposes /dev/kvm on its hosted VMs, so Android emulators and nested VMs need boat.

Can I SSH into either one?

No. CodeSandbox SDK connects through its SDK with terminals, commands and files, and smol cloud uses a command API and a WebSocket terminal capped at one hour per connection. Every boat sandbox has SSH as well as an exec and files API.

Can I self-host either one?

smol machines, yes: the smolvm engine and SDK are open source and run on macOS, Linux and Windows hosts. CodeSandbox SDK does not offer self-hosting, and boat is a hosted service only.

Is CodeSandbox SDK still maintained?

CodeSandbox is now part of Together AI. Together's docs call the CodeSandbox-based SDK deprecated and name together-sandbox as its successor, while self-serve sign-up still runs through CodeSandbox.

Try boat

Install the CLI, then run boat new.

7-day free trial

More comparisons