CodeSandbox SDK vs Deno Sandbox

Both run code in Firecracker microVMs. CodeSandbox SDK keeps a VM around, hibernating it with its memory and waking it on the next request, while Deno Sandbox is built for short, locked-down runs of up to 30 minutes that start in under 200 ms.

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

Deno Sandbox

Deno Sandbox is part of Deno Deploy, the hosting platform from the company behind the Deno runtime, and runs untrusted or AI-generated code in Linux microVMs.

Each sandbox is a Firecracker microVM with 2 vCPU, up to 4 GB of memory and a 10 GB disk that is wiped when it ends, created from a TypeScript or Python SDK. Outbound traffic goes through a proxy that can enforce a host allowlist and swap placeholder secrets for real ones. It is built for short runs of agent or user code, and a sandbox can be promoted to a Deno Deploy app.

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 rankDeno Sandbox: price rankboat: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
Hours of installs, builds and Docker outlast Deno's 30-minute lifetime and 2 vCPU shape; CodeSandbox SDK has no hard cap, and boat adds SSH, /dev/kvm and coding agents built in.
178th of 367, with limits
$1,939 a month
290th of 367, with limits
$5,665 a month
3rd cheapest of 367
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
An assistant that lives on its machine all month needs one that stays up, which CodeSandbox SDK and boat allow and a 30-minute Deno sandbox does not.
215th of 366, with limits
$2,478 a month
287th of 366, with limits
$5,803 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 suit short code runs: CodeSandbox SDK wakes a hibernated VM with its memory, Deno starts a fresh one in under 200 ms and bills CPU only while code runs.
46th cheapest of 345
$642 a month
141st of 345, with limits
$329 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
CodeSandbox SDK wakes a user's sandbox on their next HTTP request and allows 250 at once on Scale; Deno Builder allows 20 at once per region.
140th of 361, with limits
$1,707 a month
148th of 361, with limits
$1,863 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
CodeSandbox SDK can fork a running VM sharing its memory, so many rollouts branch from one warm state; Deno has no fork of a running sandbox.
348th of 360, with limits
$3,861 a month
122nd of 360, with limits
$7,963 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
296th of 364, with limits
$1,563 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
CI jobs that build Docker images or Android apps need Docker and /dev/kvm; CodeSandbox SDK runs Docker, Deno does not document it, and only boat exposes /dev/kvm.
92nd cheapest of 377
$860 a month
304th of 377, with limits
$731 a month
4th cheapest of 377
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A dev box that stays on all month needs a disk that persists; CodeSandbox SDK and boat keep it, while Deno wipes the root disk when a sandbox ends.
310th of 368, with limits
$240 a month
179th of 368, with limits
$200 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 Deno Sandbox has a desktop; boat streams a Linux desktop at 60fps with Chromium for computer use.
206th of 369, with limits
$691 a month
280th of 369, with limits
$1,583 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
251st of 367, with limits
$443 a month
284th of 367, with limits
$783 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
286th of 346, with limits
$573 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
247th of 338, with limits
$11,923 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
251st of 360, with limits
$367 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 SDKDeno Sandboxboat
IsolationFirecracker microVM with its own kernelFirecracker microVM with its own kernelFull VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
Machine size2 to 16 cores, up to 32 GB on Scale; up to 64 cores / 128 GB on Enterprise2 vCPU fixed, 768 MB to 4 GB memory, 10 GB disk4 vCPU / 8 GB by default, up to 16 vCPU / 32 GB on the $100 plan and above
Session lengthNo hard cap; hibernates after an idle timeout of up to one dayUp to 30 minutes; longer work can be promoted to a Deno Deploy appNo cap on paid plans; you set the time to live
Disk that persistsDisk kept across hibernation; 20 GB per VMRoot disk wiped when the sandbox ends; opt-in volumes of 300 MB to 20 GB, Chicago onlyThe whole disk is kept, saved every minute and on stop; a stopped sandbox costs nothing
Pause and resumeHibernates with a memory snapshot; resumes with processes still runningNo pause; a sandbox ends with its sessionDisk saved on stop; resume boots the machine again
ForkFork a hibernated sandbox from its snapshot, or a running one sharing its memoryRead-only volume snapshots that new sandboxes boot from; no fork of a running sandboxFork a running sandbox with its disk
Start timeWakes from hibernation in seconds; an HTTP request can wake itUnder 200 msBoots a full VM
Docker and nested virtualizationDocker and root, built from any Dockerfile; nested virtualization not documentedRoot access; Docker and nested virtualization not documentedDocker, systemd and /dev/kvm for Android emulators and VMs
Shell accessTerminals, commands and files through the SDK; no SSHSSH and a VS Code editor in the browserSSH, plus exec and files over the API
Inbound accessHTTPS host per port, public by default, signed tokens for private hostsHTTPS URL per exposed port, public without authenticationHTTPS URLs for up to 50 ports, token-protected by default; raw TCP
Outbound rulesNo documented host allowlistHost allowlist per sandbox, enforced by a proxy outside the VMOpen internet, no provider-side allowlist
Keeping API keys out of the sandboxSecrets passed in as environment variablesSecrets swapped in on outbound requests to approved hosts; the sandbox only sees a placeholderCredentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out
Desktop and coding agentsNo desktop; run your own agent through the command APINo desktop; a root image with Claude Code, run through shell commandsLinux desktop at 60fps with Chromium; seven agents preinstalled with a prompt, steer and events API
RegionsA country hint steers scheduling, for example to Europe, but placement is not guaranteedAmsterdam and ChicagoEU only (Germany, Finland, France)
Every feature Battleships tracks (43)
FeatureCodeSandbox SDKDeno Sandboxboat
Full VM (own kernel)YesYesYes
Docker insideYesNot publishedYes
Nested virtualizationNoNoYes
RootYesYesYes
SystemdNot publishedNot publishedYes
Desktop GUINoNoYes
BrowserNoNoYes
Computer-use APINoNoYes
SSHNoYesYes
Persistent diskYesYesYes
Pause/resumeYesNoYes
Automatic snapshotsYesNoYes
Fork/cloneYesNoYes
Live fork (no pause)YesNoYes
Memory forkYesNoNo
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesNoYes
Idle auto-stopYesNoNo
Wake on requestYesNoNo
HTTPS preview URLsYesYesYes
Raw TCP inboundNoNoYes
Public IPv4NoNoNo
Egress allowlistNoYesNo
Secret proxyNoYesYes
Secret proxy for any APINoYesYes
Preinstalled agentsNoYesYes
Agent harness APINoNoYes
MCP serverNoNoNo
GPUNoNoNo
WindowsNoNoNo
Arm64NoNoNo
Self-hostingNoNoNo
BYOCNoNoNo
Open sourceNoNoNo
Scoped API keysNot publishedNot publishedYes
Spend limitsNot publishedYesYes
Usage per sandbox (API)NoNoYes
Audit logsNot publishedNot publishedNo
SSOYesNoNo
SOC 2YesYesNo
HIPAANoYesNo
EU data residencyNoYesYes
Zero data retentionNot publishedNot publishedYes

Performance

CodeSandbox SDKNot measured
Deno SandboxNot measured
boat26.69 runs/s

There is no comparable CPU measurement for CodeSandbox SDK or Deno Sandbox. boat did 26.69 runs per second on the StarSling hpc-sandbox-benchmarks Node.js loop at 4 vCPU / 8 GB, on dedicated AMD Ryzen 9 9950X hosts; neither of the other two was in that run, and Deno's fixed 2 vCPU shape could not run the 4 vCPU test as set up.

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 Deno Sandbox 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 monthDeno Sandbox, 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$53 · Pro · with limits$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$341 · Scale$200 · Builder · with limits$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$2,779 · Builder · with limits$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$783 · Builder · with limits$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$1,563 · Builder · with limits$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. Teams pay a $170 monthly Scale platform fee on top, which is not usage credit, and a hibernated VM costs nothing.

Deno Sandbox: Sandboxes need Pro at $20 a month or Builder at $200 a month; the fee is not credit but includes CPU, memory and egress allowances shared with Deno Deploy apps. Past those, CPU is $0.10 per hour of active CPU and memory $0.025 per GiB-hour for the whole life of the sandbox.

Deno Sandbox: 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

Deno Sandbox fits none of the five scenarios as written: a sandbox lives up to 30 minutes, is fixed at 2 vCPU and 4 GB, and Builder allows 20 at once per region. CodeSandbox SDK fits the workday agent, the always-on agent and 50 agents in parallel, but its Scale plan caps at 250 VMs at once, so the user-sandboxes and 1,000-job scenarios need Enterprise, and its disk past 20 GB and egress are not priced. boat fits four of the five; at 1,000 short jobs at once its top plan queues starts, so none of the three fits that one without a custom plan.

ScenarioCodeSandbox SDKDeno Sandboxboat
One coding agent, workdays
  • disk beyond 20 GiB: price unknown
  • egress price not published
  • Pro: sessions capped at 0.5 h
  • capped at 2 vCPU / 4 GiB
  • largest size is 2 vCPU / 4 GiB
  • disk beyond 10 GiB: price unknown
None that bite
One agent that never stops
  • disk beyond 20 GiB: price unknown
  • egress price not published
  • Builder: sessions capped at 0.5 h
  • capped at 2 vCPU / 4 GiB
  • largest size is 2 vCPU / 4 GiB
  • disk beyond 10 GiB: price unknown
None that bite
50 agents in parallel
  • egress price not published
  • Builder: max 20 concurrent
  • Builder: sessions capped at 0.5 h
  • capped at 2 vCPU / 4 GiB
  • largest size is 2 vCPU / 4 GiB
None that bite
Sandboxes for your users
  • Scale: max 250 concurrent
  • egress price not published
  • Builder: max 20 concurrent
None that bite
1,000 short jobs at once
  • Scale: max 250 concurrent
  • egress price not published
  • Builder: max 20 concurrent
  • 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, or need more than 2 vCPU for longer than 30 minutes, and you are fine building on what Together now calls its legacy SDK.

  • VMs that sleep and wake with their memory. Useful for previews and code runs used in bursts: an HTTP request wakes a hibernated VM with its processes still running.
  • No 30-minute ceiling and bigger machines. Fits longer jobs: a VM runs until it goes idle, with up to 16 cores on Scale and 64 on Enterprise.
  • Forks that share memory. Good for branching many runs, such as rollouts or test variants, from one warm running state.

Use Deno Sandbox when

you run short pieces of untrusted code that must start in milliseconds, reach only the hosts you allow, or run in the US, especially if you already deploy on Deno Deploy.

  • Locked-down outbound traffic. Useful for untrusted code: each sandbox can be limited to a list of hosts, and secrets stay outside the VM.
  • Starts in under 200 ms. Good for one sandbox per request: it starts fast and CPU bills only while code is running.
  • A US region and Deno Deploy. Fits teams on Deno Deploy who want Chicago or Amsterdam and a path from sandbox to a live app.

Where boat fits

If what you are really running is agents that work for hours, look at boat. CodeSandbox SDK and Deno Sandbox are both good at short code runs, but neither gives an agent a full machine with SSH, nested virtualization, a desktop and its coding harness already set up.

A full machine with KVM and SSH

Neither documents /dev/kvm, so Android emulators and VM-based tests cannot start. CodeSandbox SDK has no SSH, and Deno is fixed at 2 vCPU and 4 GB, which big builds outgrow. boat runs Docker, systemd and /dev/kvm on up to 16 vCPU.

A disk that stays, without a clock

A Deno sandbox ends after 30 minutes and wipes its root disk, so long tasks must be split and rebuilt. CodeSandbox SDK keeps the disk but bills the idle time before it hibernates. boat saves the disk every minute and a stopped sandbox costs nothing.

Agents, desktop and hosting built in

On both you build the agent harness and browser yourself, and hosts are public by default. boat ships seven agents you drive from outside, a 60fps desktop with Chromium, and token-protected HTTPS on 50 ports plus raw TCP.

Use boat when your agents do real work for hours: installs, builds, Docker, emulators, a browser and servers other people reach, on a full machine whose disk is still there tomorrow and whose agent and keys are already set up. Compare it one to one: boat vs CodeSandbox SDK and boat vs Deno Sandbox.

Questions people ask

Is CodeSandbox SDK cheaper than Deno Sandbox?

Deno often looks cheaper, but it fits none of the five scenarios as written. For 50 agents in parallel, CodeSandbox SDK comes to $2,500 a month and fits, while Deno Builder comes to $2,779 a month and caps at 20 at once and 30 minutes each. boat fits that one at $317 a month.

Which is faster, CodeSandbox SDK or Deno Sandbox?

Neither has a comparable CPU measurement, so there is no fair speed ratio. Deno says a sandbox starts in under 200 ms, while CodeSandbox SDK resumes a hibernated VM from a memory snapshot in seconds with processes still running.

Can CodeSandbox SDK or Deno Sandbox run Docker?

CodeSandbox SDK runs Docker with root and builds sandboxes from any Dockerfile. Deno's docs give root access but do not say Docker works, and neither documents nested virtualization. boat runs Docker and systemd and exposes /dev/kvm.

How long can a sandbox run on each?

A Deno sandbox lives up to 30 minutes, and longer work can be promoted to a Deno Deploy app. CodeSandbox SDK has no hard cap and hibernates after an idle timeout of up to one day. boat has no cap on paid plans.

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.

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.

Try boat

Install the CLI, then run boat new.

7-day free trial

More comparisons