CodeSandbox SDK vs Islo

Both give each sandbox a microVM with its own kernel. CodeSandbox SDK is built around VMs that hibernate with their memory and wake on the next HTTP request, while Islo is built around coding agents whose outbound traffic passes a gateway that allows, blocks, injects keys and logs.

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

Islo

Islo is an agent sandbox from Incredibuild: each coding agent gets its own microVM, with network rules and credentials handled at a gateway.

An Islo sandbox is a Linux microVM with its own kernel, built from a container image, with Claude Code, Cursor CLI and Codex preinstalled. Outbound traffic can pass through a gateway that allows or blocks hosts, injects credentials and logs every request. It is built for teams that run coding agents, evals and RL rollouts in the background, on Islo's cloud or in their own cloud account.

Launched
May 4, 2026

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 rankIslo: 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 their microVMs, but neither documents /dev/kvm, so Android emulators and nested VMs need boat.
178th of 367, with limits
$1,939 a month
326th of 367, with limits
$7,402 a month
3rd cheapest of 367
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
Islo fits an assistant that lives on its machine, with preinstalled agents and pause on idle, while CodeSandbox SDK leaves the agent harness to you.
215th of 366, with limits
$2,478 a month
30th cheapest of 366
$9,928 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
CodeSandbox SDK fits short code runs well, since VMs hibernate with memory and wake on the next request, while Islo needs a raised concurrency limit for big bursts.
46th cheapest of 345
$642 a month
153rd of 345, with limits
$503 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
300 user sandboxes at once is more than CodeSandbox SDK's 250 on Scale and Islo's reported 50 on Team, so both need a custom limit; boat fits it.
140th of 361, with limits
$1,707 a month
171st of 361, with limits
$3,160 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's memory forks let rollouts branch from one warm state, while Islo snapshots files and starts new sandboxes from them.
348th of 360, with limits
$3,861 a month
167th of 360, with limits
$15,265 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
CodeSandbox SDK runs evals on up to 250 VMs at once on Scale, Islo adds a Harbor integration and gateway rules that stop trials looking up answers, and boat runs more sandboxes at once than either.
64th cheapest of 364
$1,335 a month
198th of 364, with limits
$2,456 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
92nd cheapest of 377
$860 a month
119th cheapest of 377
$1,507 a month
4th cheapest of 377
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
Both keep a VM running with no session cap, but neither gives you a desktop or /dev/kvm, which boat has on every machine.
310th of 368, with limits
$240 a month
112th cheapest of 368
$402 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 has a desktop or a computer-use API, while boat streams a 60fps desktop with Chromium that agents can see and click.
206th of 369, with limits
$691 a month
310th of 369, with limits
$2,210 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
113th of 367, with limits
$1,105 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
66th cheapest of 346
$748 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
265th of 338, with limits
$14,966 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
47th of 360, with limits
$746 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 SDKIsloboat
IsolationFirecracker microVM with its own kernelMicroVM with its own kernel, built from a container imageFull VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
Docker and nested virtualizationDocker and root, built from any Dockerfile; no /dev/kvm documentedDocker inside and sudo; no /dev/kvm or nested virtualization documentedDocker, root, systemd and /dev/kvm on every machine
Pause and resumeHibernates with a memory snapshot; an HTTP request to a host wakes it with processes still runningPause keeps files, processes and memory; resumes on the next exec, file access or webhookDisk saved every minute and on stop; a stopped sandbox costs nothing; resume boots it again
ForkFork a hibernated sandbox from its memory snapshot, or a running one sharing its memorySnapshot a running sandbox's files, then start new sandboxes from itFork a running sandbox with its disk
Shell accessTerminals, commands and files through the SDK; no SSHSSH and port-forward through the Islo CLISSH, 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 IPHTTPS share links anyone with the URL can open, expiring after a TTL (24 hours by default); no raw TCPHTTPS URLs for up to 50 ports, token-protected by default; raw TCP; its own IP
Desktop and browserNo desktop; a headless Chromium templateA browser inside the sandbox for testing; no desktop or computer-use APILinux desktop streamed at 60fps, Chromium and a computer-use API
Coding agentsRun your own agent through the command APIClaude Code, Cursor CLI and Codex preinstalled; factory lines run agent steps on a schedule, webhook or GitHub, Slack and Linear eventsClaude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API
Outbound internet rulesNo egress allowlist documentedGateway profiles allow or deny by host, path and HTTP method, with default deny, rate limits and content filtersOpen internet on every plan; no provider-side allowlist
Keeping API keys out of the sandboxSecrets passed in as environment variablesGateway injects stored credentials into outbound requests; keys never reach the sandboxCredentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out
Request logsNo per-request outbound log documentedPer-request gateway logs of what each agent called and what was blocked, streamed to your SIEMUsage API and spend limits; no per-request log of outbound calls
Sandboxes at once10 on Build, 250 on Scale, more on Enterprise5 on Free and 50 on Team per launch coverage; paused ones do not count; more by contract100 to 2,000 running sandboxes by plan, raised by adding seats
Where it runsNo guaranteed region; a country hint can steer to a cluster in EuropeUS West, or inside your own cloud account with BYOCEU only: Germany, Finland and France, with EU data residency
Every feature Battleships tracks (43)
FeatureCodeSandbox SDKIsloboat
Full VM (own kernel)YesYesYes
Docker insideYesYesYes
Nested virtualizationNoNoYes
RootYesYesYes
SystemdNot publishedNot publishedYes
Desktop GUINoNoYes
BrowserNoYesYes
Computer-use APINoNoYes
SSHNoYesYes
Persistent diskYesYesYes
Pause/resumeYesYesYes
Automatic snapshotsYesYesYes
Fork/cloneYesYesYes
Live fork (no pause)YesYesYes
Memory forkYesNoNo
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesYesYes
Idle auto-stopYesYesNo
Wake on requestYesYesNo
HTTPS preview URLsYesYesYes
Raw TCP inboundNoNoYes
Public IPv4NoNoNo
Egress allowlistNoYesNo
Secret proxyNoYesYes
Secret proxy for any APINoYesYes
Preinstalled agentsNoYesYes
Agent harness APINoYesYes
MCP serverNoNoNo
GPUNoNoNo
WindowsNoNoNo
Arm64NoNoNo
Self-hostingNoNoNo
BYOCNoYesNo
Open sourceNoNoNo
Scoped API keysNot publishedNot publishedYes
Spend limitsNot publishedNot publishedYes
Usage per sandbox (API)NoYesYes
Audit logsNot publishedYesNo
SSOYesYesNo
SOC 2YesNot publishedNo
HIPAANoNot publishedNo
EU data residencyNoNoYes
Zero data retentionNot publishedNot publishedYes

Performance

CodeSandbox SDKNot measured
IsloNot measured
boat26.69 runs/s

There is no comparable CPU measurement for CodeSandbox SDK or Islo: neither was part of StarSling's hpc-sandbox-benchmarks run. In that run of 2026-09-29, boat did 26.69 Node.js runs per second at 4 vCPU / 8 GB on dedicated AMD Ryzen 9 9950X hosts.

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 Islo 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 monthIslo, 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$101 · Free$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$341 · Scale$420 · Free$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$4,863 · Team$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$1,382 · Team · 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$3,053 · Team · 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. Scale adds a $170 monthly platform fee that is not usage credit, and a hibernated VM costs nothing.

Islo: Islo bills each CPU core and each GB of memory and disk while a sandbox runs, with no plan fee. An idle sandbox keeps billing until a pause-after-idle timer fires, which can only be set at creation, and egress has no published price.

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

Limits in those scenarios

CodeSandbox SDK's Scale plan caps at 250 VMs at once and Islo's Team plan is reported at 50, so 300 sandboxes for your users and 1,000 short jobs at once fit neither without a custom contract. CodeSandbox SDK VMs have a 20 GB disk with no published price for more, and neither publishes an egress price, so their totals leave egress out. boat fits the 300 user sandboxes, but its start limits queue the 1,000-job burst even on the $2,000 plan.

ScenarioCodeSandbox SDKIsloboat
One coding agent, workdays
  • disk beyond 20 GiB: price unknown
  • egress price not published
  • egress price not published
None that bite
One agent that never stops
  • disk beyond 20 GiB: price unknown
  • egress price not published
  • egress price not published
None that bite
50 agents in parallel
  • egress price not published
  • egress price not published
None that bite
Sandboxes for your users
  • Scale: max 250 concurrent
  • egress price not published
  • Team: max 50 concurrent
  • egress price not published
None that bite
1,000 short jobs at once
  • Scale: max 250 concurrent
  • egress price not published
  • Team: max 50 concurrent
  • egress price not published
  • 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 memory intact, or you fork many runs from one warm state, and you are fine building on what Together now calls its legacy SDK.

  • Wake on the next request. For previews and code runs used in bursts: a hibernated VM wakes on an HTTP request with its memory and processes intact.
  • Forks that share memory. For branching many runs from one warm state, such as rollouts or test variants, without booting and setting up again.
  • More sandboxes at once. For teams that need many VMs together: Scale allows 250 at once, against the 50 reported for Islo's Team plan.

Use Islo when

you run background coding agents and need per-host egress rules, keys injected at a gateway and a log of every outbound call, in US West or your own cloud account.

  • Egress rules and request logs. For teams that must control what agents reach: Islo allows or blocks each host, path and method, and logs every call.
  • Keys stay at the gateway. For agents that call outside APIs: Islo injects stored credentials on the way out, so keys never sit inside the sandbox.
  • Coding agents built in. For teams running background agents: Claude Code, Cursor CLI and Codex come preinstalled, with triggers from schedules, webhooks, GitHub, Slack and Linear.

Where boat fits

If your agents do real work for hours, with builds, emulators, a browser on a screen and servers other people reach, look at boat. It gives each one a full VM with SSH, /dev/kvm, a 60fps desktop and hosting on 50 ports, with coding agents and a credentials proxy built in, and runs up to 2,000 sandboxes at once.

Emulators and nested VMs run

Neither CodeSandbox SDK nor Islo documents /dev/kvm, so Android emulators and tools that start their own VMs stop at the first boot. Software emulation, where it works at all, runs many times slower.

A screen agents can use

Both are headless: no desktop and no computer-use API. An agent that must check a UI or fill a form in a real app needs you to install a VNC stack and a control layer in every image.

Ports that stay reachable

CodeSandbox SDK hosts are HTTPS only, and Islo share links are public to anyone with the URL and expire after 24 hours by default. Neither takes raw TCP, so a database or game server an agent starts cannot be reached.

Use boat when your agents need a whole machine: SSH, Docker with /dev/kvm, a 60fps desktop for computer use, ports others can reach and up to 2,000 sandboxes at once, in the EU. Compare it one to one: boat vs CodeSandbox SDK and boat vs Islo.

Questions people ask

Is CodeSandbox SDK cheaper than Islo?

It depends on the shape. One coding agent on workdays comes to $193 a month on CodeSandbox SDK and $101 on Islo, but one agent running all month is $341 against $420, and 50 agents in parallel are $2,500 against $4,863. boat comes to $20, $26 and $317 for the same three.

Which is faster, CodeSandbox SDK or Islo?

We have no comparable CPU measurement for either, since neither was in StarSling's public benchmark run. boat did 26.69 runs per second in that run at 4 vCPU / 8 GB.

Can CodeSandbox SDK and Islo run Docker?

Yes, both run Docker inside their microVMs with root or sudo. Neither documents /dev/kvm, so Android emulators and nested VMs are out; boat has Docker, systemd and /dev/kvm on every machine.

Can I SSH into CodeSandbox SDK or Islo?

Islo gives SSH and port-forwarding through its CLI. CodeSandbox SDK connects through its SDK, Node and browser clients with terminals, commands and files, and no SSH. Every boat sandbox has SSH as well as an exec and files API.

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.

How many sandboxes can run at once?

CodeSandbox SDK allows 250 VMs at once on Scale. Islo's Team plan is reported at 50 running sandboxes, with paused ones not counted. boat runs 100 to 2,000 by plan.

Try boat

Install the CLI, then run boat new.

7-day free trial

More comparisons