Vercel Sandbox vs CodeSandbox SDK
Both run each sandbox in a Firecracker microVM with its own kernel. Vercel Sandbox is built for short runs of untrusted code in 19 regions, with sessions that stop after 24 hours, while CodeSandbox SDK keeps VMs that hibernate with their memory and wake on the next request.
The companies and products
Vercel Sandbox
Vercel is a cloud platform for web apps and agents, and Vercel Sandbox is its product for running untrusted or agent-written code in isolated Linux microVMs.
Each sandbox is a Firecracker microVM with its own kernel, started from a Vercel image, your own OCI image, or a saved snapshot. You drive it with the JavaScript or Python SDK or the sandbox CLI: run commands, move files, open ports, and set network rules. It is aimed at code interpreters, app builders, test runs and agent tools, and Vercel's docs say it is not meant to run nonstop.
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.
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. “22nd cheapest of 345” means 21 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 workload | Vercel Sandbox: price rank | CodeSandbox SDK: price rank | boat: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month | 244th of 367, with limits $3,351 a month | 178th of 367, with limits $1,939 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An agent that never stops fits CodeSandbox SDK better than Vercel, whose sessions end at 24 hours; boat runs with no session cap and keeps its disk. | 257th of 366, with limits $3,657 a month | 215th of 366, with limits $2,478 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 bursts; Vercel bills CPU only while code uses it, and CodeSandbox SDK wakes hibernated VMs with their memory on the next request. | 22nd cheapest of 345 $216 a month | 46th cheapest of 345 $642 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 Vercel ranks 15th here and fits 300 sandboxes at once, while CodeSandbox SDK's Scale plan stops at 250 VMs; boat ranks first on cost for the same work. | 15th cheapest of 361 $1,267 a month | 140th of 361, with limits $1,707 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 Vercel ranks 3rd thanks to high concurrency and fast starts; CodeSandbox SDK's memory forks help, but its 250 VM cap on Scale holds it back. | 3rd cheapest of 360 $15,761 a month | 348th of 360, with limits $3,861 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 | 69th cheapest of 364 $1,508 a month | 64th cheapest of 364 $1,335 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 Both run Docker for CI, but jobs that build Android apps or VM images need /dev/kvm, which boat has and neither of the other two offers. | 107th cheapest of 377 $1,144 a month | 92nd cheapest of 377 $860 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Vercel restarts a dev box every 24 hours, and CodeSandbox SDK bills idle time until it hibernates; boat stays up all day and costs nothing when stopped. | 276th of 368, with limits $145 a month | 310th of 368, with limits $240 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 Vercel Sandbox nor CodeSandbox SDK has a desktop, so a computer-use agent needs one built by hand; boat streams a 60fps desktop with a computer-use API. | 235th of 369, with limits $952 a month | 206th of 369, with limits $691 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 Vercel has no browser feature and CodeSandbox SDK offers a headless Chromium template; boat ships Chromium on a full desktop with browser control. | 259th of 367, with limits $486 a month | 251st of 367, with limits $443 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 | 277th of 346, with limits $468 a month | 281st of 346, with limits $508 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 | 234th of 338, with limits $9,960 a month | 193rd of 338, with limits $7,390 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 | 283rd of 360, with limits $554 a month | 240th of 360, with limits $326 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 it | Can't run it | Can't run it |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Can't run it | Can't run it | Can't run it |
Key features
| Feature | Vercel Sandbox | CodeSandbox SDK | boat |
|---|---|---|---|
| Isolation | Firecracker microVM with its own kernel | Firecracker microVM with its own kernel | Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts |
| Longest session | 24 hours on Pro and Enterprise, 45 minutes on Hobby | No hard cap; hibernates after an idle timeout of up to one day | No cap on paid plans; you set the time to live |
| Pause and resume | Filesystem snapshot on stop; a new session boots from it and processes start over | Hibernates with a memory snapshot; resumes with processes still running | Disk saved every minute and on stop; resume boots the machine again; a stopped sandbox costs nothing |
| Wake on request | No | Yes; an HTTP request to a host wakes a hibernated VM | No; a stopped sandbox is started by API |
| Fork | Fork from a filesystem snapshot or a running sandbox; no memory fork | Fork a hibernated sandbox from its memory snapshot, or a running one sharing its memory | Fork a running sandbox with its disk |
| Biggest machine | 8 vCPU / 16 GB on Pro, 32 vCPU / 64 GB on Enterprise | 16 cores / 32 GB on Scale, 64 cores / 128 GB on Enterprise | Up to 16 vCPU / 32 GB on the $100 plan and above |
| Regions | 19 regions across the US, Europe, Asia, South America, Africa and Australia | A country hint steers scheduling, but placement is not guaranteed | EU only: Germany, Finland, France |
| Egress rules | Allow-all, deny-all or a domain and CIDR allowlist, changeable while running | No provider-side allowlist documented | No provider-side allowlist; firewall inside the VM |
| Docker inside | Yes, with sudo; images from your own Dockerfile | Yes, with root; built from any Dockerfile | Yes, with root and systemd; named snapshots as templates, no Dockerfile builds |
| Nested virtualization | No | Not documented | Yes, /dev/kvm for Android emulators and VMs |
| Inbound access | Public URLs for up to 15 ports, no built-in auth; no raw TCP or public IP | HTTPS host per port, public by default, signed tokens for private hosts; no raw TCP or public IP | HTTPS URLs for up to 50 ports, token-protected by default; raw TCP; dedicated IPv6 or IPv4 |
| Desktop and browser | None; headless Linux only | No desktop; a headless Chromium template | 60fps desktop with Chromium, computer-use and browser-control APIs |
| Coding agents | Claude Code, Codex, OpenCode and pi preinstalled; you start them as commands | Run your own agent through the command API | Claude Code, Codex and others preinstalled, with an API to prompt, steer and stream events |
| Keeping API keys out of the sandbox | Credentials brokering injects headers for any HTTPS domain you allow | Secrets passed in as environment variables | Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out |
Every feature Battleships tracks (43)
| Feature | Vercel Sandbox | CodeSandbox SDK | boat |
|---|---|---|---|
| Full VM (own kernel) | Yes | Yes | Yes |
| Docker inside | Yes | Yes | Yes |
| Nested virtualization | No | No | Yes |
| Root | Yes | Yes | Yes |
| Systemd | Not published | Not published | Yes |
| Desktop GUI | No | No | Yes |
| Browser | No | No | Yes |
| Computer-use API | No | No | Yes |
| SSH | No | No | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | Yes | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Yes | Yes | Yes |
| Memory fork | No | Yes | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | No | Yes | No |
| Wake on request | No | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | No | No | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | Yes | No | No |
| Secret proxy | Yes | No | Yes |
| Secret proxy for any API | Yes | No | Yes |
| Preinstalled agents | Yes | No | Yes |
| Agent harness API | No | No | Yes |
| MCP server | Yes | No | No |
| GPU | No | No | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | No | No | No |
| BYOC | No | No | No |
| Open source | No | No | No |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | Yes | Not published | Yes |
| Usage per sandbox (API) | No | No | Yes |
| Audit logs | Not published | Not published | No |
| SSO | Yes | Yes | No |
| SOC 2 | Yes | Yes | No |
| HIPAA | Yes | No | No |
| EU data residency | Yes | No | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
On the StarSling hpc-sandbox-benchmarks Node.js loop at 4 vCPU / 8 GB, Vercel Sandbox did 12.96 runs per second and boat did 26.69, so boat finished about twice as many runs in the same time. CodeSandbox SDK has no comparable CPU measurement, so we cannot say how it compares on 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 has 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.
| Scenario | Vercel Sandbox, a month | CodeSandbox SDK, a month | boat, 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 | $57 · Pro | $193 · Scale | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $236 · Pro · with limits | $341 · Scale | $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 | $2,884 · Pro | $2,500 · Scale | $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 | $828 · Pro | $815 · Scale · 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 | $3,152 · Pro | $1,632 · Scale · with limits | $2,000 · $2000/mo · with limits |
Vercel Sandbox: Vercel charges for CPU only while your code uses it, but charges for memory for the whole time the sandbox runs, plus a small fee per sandbox created and a monthly fee per GB of snapshots. The $20 Pro credit is shared with your other Vercel usage.
CodeSandbox SDK: You buy VM credits and each size burns a fixed number per hour, billed per started minute while the VM runs, including idle time before it hibernates. The Scale plan adds a $170 monthly platform fee that is not usage credit, and a hibernated VM costs nothing.
CodeSandbox SDK: boat charges a flat price per size, per second, only while the machine runs, and a stopped sandbox costs nothing. Plans start at $20 a month, and all of the fee is usage credit.
Limits in those scenarios
For one agent that never stops, Vercel ends each session at 24 hours and running processes start over, while CodeSandbox SDK keeps going. CodeSandbox SDK's Scale plan allows 250 VMs at once, so 300 user sandboxes and 1,000 short jobs do not fit it without Enterprise, and its price for disk beyond 20 GiB and for egress is not published. boat fits every scenario but the 1,000-job burst, where even the $2,000 plan queues starts.
| Scenario | Vercel Sandbox | CodeSandbox SDK | boat |
|---|---|---|---|
| One coding agent, workdays | None that bite |
| None that bite |
| One agent that never stops |
|
| None that bite |
| 50 agents in parallel | None that bite |
| None that bite |
| Sandboxes for your users | None that bite |
| None that bite |
| 1,000 short jobs at once | None that bite |
|
|
Which one should you use?
Use Vercel Sandbox when
you run many short bursts of untrusted code, want egress rules you change from outside, or need to run close to users outside Europe.
- Regions all over the world. Useful if your users are spread out: Vercel Sandbox runs in 19 regions, while CodeSandbox SDK only takes a placement hint.
- A firewall you control from outside. A running sandbox can be switched to deny-all, or to a list of allowed domains, without a restart.
- Room for big bursts. Useful for thousands of short jobs: Pro allows far more sandboxes at once than CodeSandbox SDK's 250 VM cap on Scale.
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 and wake on request. Useful for sandboxes that sleep between visits: an HTTP request wakes a hibernated VM with its processes still running.
- Forks that share memory. Good for branching many runs from one warm state, such as rollouts or test variants.
- No 24-hour session cap. Useful for agents that keep working: a busy VM keeps running, where Vercel stops every session after 24 hours.
Where boat fits
Both Vercel Sandbox and CodeSandbox SDK are headless code boxes you drive through an SDK. If your agents work for hours on one machine, with builds, Docker, emulators, a browser and servers other people reach, boat gives each one a full VM with the agent and its keys already set up.
A whole machine with SSH and KVM
Neither offers SSH, and Vercel has no nested virtualization while CodeSandbox SDK does not document it. Android emulator builds and VM-based tests cannot start, and every tool goes through an SDK call instead of a shell.
A desktop and a browser built in
Neither has a desktop, so an agent cannot open the web app it built to see if it works. On Vercel you build the desktop yourself; on CodeSandbox SDK you start from a headless Chromium template and add the rest.
Drive the agents inside from outside
Vercel installs coding agents but you run them as plain commands, and on CodeSandbox SDK you bring your own. Either way you write the code that prompts, steers and reads every agent; boat gives you one API for that.
Use boat when your agents work for hours on one machine: builds, Docker, emulators, a desktop with a browser and servers other people reach, with an API to drive the agents inside from outside. Compare it one to one: boat vs Vercel Sandbox and boat vs CodeSandbox SDK.
Questions people ask
Is Vercel Sandbox cheaper than CodeSandbox SDK?
It depends on the job. One workday agent costs $57 on Vercel vs $193 on CodeSandbox SDK, but 50 parallel agents cost $2,884 vs $2,500. For 1,000 short jobs at once Vercel fits at $3,152 a month, while CodeSandbox SDK's Scale plan caps at 250 VMs.
Which is faster, Vercel Sandbox or CodeSandbox SDK?
Vercel Sandbox did 12.96 runs per second on the hpc-sandbox-benchmarks Node.js loop at 4 vCPU / 8 GB. CodeSandbox SDK was not part of that run, so there is no fair speed comparison. boat did 26.69 on the same test.
Can either one run 24/7?
Vercel Sandbox cannot as one session: Pro and Enterprise sessions stop at 24 hours, and its docs say sandboxes are not designed to run continuously. CodeSandbox SDK has no hard session cap, so a busy VM keeps running and only hibernates when idle.
Can Vercel Sandbox and CodeSandbox SDK run Docker?
Yes, both run Docker inside the sandbox and can boot from your own Dockerfile. Neither offers /dev/kvm for Android emulators or nested VMs, which boat does.
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.
Where do they run?
Vercel Sandbox runs in 19 regions, with iad1 in the US as the default. CodeSandbox SDK takes a country hint that can steer a sandbox to a cluster in Europe, but placement is not guaranteed. boat runs only in the EU.
Try boat
Install the CLI, then run boat new.
7-day free trial
More comparisons
- boat vs Vercel Sandbox
- boat vs CodeSandbox SDK
- E2B vs Vercel Sandbox
- Daytona vs Vercel Sandbox
- Modal vs Vercel Sandbox
- Vercel Sandbox vs Cloudflare Sandbox
- Vercel Sandbox vs Blaxel
- Vercel Sandbox vs Runloop
- Vercel Sandbox vs Tensorlake
- Vercel Sandbox vs Morph Cloud
- Vercel Sandbox vs Northflank
- Vercel Sandbox vs Fly.io Machines