Modal vs Vercel Sandbox
Modal runs sandboxes as gVisor containers on a big AI compute platform with GPUs, and a finished sandbox is gone. Vercel Sandbox runs each one as a Firecracker microVM with its own kernel and Docker, and a stopped sandbox resumes from a saved snapshot.
The companies and products
Modal
Modal is a serverless cloud for AI work: inference, training, batch jobs and sandboxed code, all driven from Python, TypeScript or Go code.
Modal Sandboxes are containers you create from code, run commands in, and then terminate. By default they run under gVisor, a user-space kernel from Google; since October 2026 you can also pick a VM runtime with its own Linux kernel for Docker and FUSE (Battleships tracks that as a separate card, Modal VM Sandboxes). It suits teams that already run models on Modal and want short-lived, scalable code execution next to their GPUs.
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.
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. “19th cheapest of 345” means 18 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 | Modal: price rank | Vercel Sandbox: price rank | boat: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month | Can't run it | 244th of 367, with limits $3,351 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Both Modal and Vercel Sandbox cap a session at 24 hours, so an always-on agent is a compromise on either; boat has no session cap on paid plans. | 196th of 366, with limits $1,924 a month | 257th of 366, with limits $3,657 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: Modal ranks 19th and Vercel Sandbox ranks 22nd, while boat's start limits make it a compromise here. | 19th cheapest of 345 $181 a month | 22nd cheapest of 345 $216 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 Both fit sandboxes for your users, but boat ranks 1st on cost for the same work, ahead of Modal at 11th and Vercel Sandbox at 15th. | 11th cheapest of 361 $811 a month | 15th cheapest of 361 $1,267 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 Many short parallel rollouts suit both, with Vercel Sandbox ranking 3rd and Modal 5th; boat's start limits slow big bursts. | 5th cheapest of 360 $19,038 a month | 3rd cheapest of 360 $15,761 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 | 59th cheapest of 364 $1,244 a month | 69th cheapest of 364 $1,508 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 run on Vercel Sandbox and on boat, but on Modal only with the VM runtime. | 126th cheapest of 377 $1,793 a month | 107th cheapest of 377 $1,144 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 up all day restarts every 24 hours on both; boat keeps it running and saves its disk every minute. | 138th of 368, with limits $23 a month | 276th of 368, with limits $145 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 Modal nor Vercel Sandbox has a desktop, so you build one yourself; boat ships a 60fps desktop with a computer-use API. | 98th of 369, with limits $183 a month | 235th of 369, with limits $952 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 | 177th of 367, with limits $136 a month | 259th of 367, with limits $486 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 | 267th of 346, with limits $394 a month | 277th of 346, with limits $468 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 | 223rd of 338, with limits $8,795 a month | 234th of 338, with limits $9,960 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 | Can't run it | 283rd of 360, with limits $554 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 Modal has GPUs from T4 to B300, and neither Vercel Sandbox nor boat has any, so Modal is the pick for GPU jobs. | 19th cheapest of 98 $1,187 a month | 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 | Modal | Vercel Sandbox | boat |
|---|---|---|---|
| Isolation | gVisor container by default; VM runtime with its own kernel available | Firecracker microVM with its own kernel | Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts |
| Docker inside | Only with the VM runtime | Yes, with sudo | Yes, with root and systemd |
| Nested virtualization | VM runtime only, Team and Enterprise plans, on request | No | Yes, /dev/kvm (Android emulator, VMs) |
| Longest session | 24 hours (default 5 minutes) | 24 hours on Pro and Enterprise, 45 minutes on Hobby | No cap on paid plans; default TTL 1 hour, changeable |
| Stop and resume | No paused state; a finished sandbox is gone, only Volumes keep files | Filesystem snapshot on stop; a new session boots from it, kept 30 days after last use | Disk saved every minute and on stop; a stopped sandbox costs nothing |
| Memory snapshots | Yes, in alpha: snapshotting ends the sandbox, snapshots expire after 7 days | No; processes start over on resume | No; the machine reboots with its disk |
| GPUs | T4 to B300; 10 at once on Starter, 50 on Team | None | None |
| Desktop, browser, computer use | Not offered; an example shows how to set up VNC yourself | None; headless Linux only | 60fps desktop and Chromium, computer-use and browser-control APIs |
| Custom images | Build images in code or pull any linux/amd64 registry image | Your own OCI images from a Dockerfile, via Vercel Container Registry | Named snapshots used as templates; no Dockerfile builds |
| Coding agents | Examples only | Claude Code, Codex, OpenCode and pi preinstalled; you start them as commands | Claude Code, Codex and five other agents, with an API to prompt, steer and stream events |
| Keys kept outside the sandbox | Secrets are environment variables, or you run your own proxy in a sidecar | Credentials brokering injects headers for any HTTPS domain you allow | Credentials proxy for model providers and any HTTPS API you register |
| Inbound access | Encrypted and raw TCP tunnels, connect tokens, inbound IP allowlist | Public URLs for up to 15 ports, no built-in auth; no raw TCP | HTTPS URLs for up to 50 ports, token-protected by default, plus raw TCP |
| Outbound network rules | Block all, IP-range allowlist, domain allowlist (beta) | Allow-all, deny-all or a domain and CIDR allowlist, changeable while running | No provider-side allowlist; firewall inside the VM |
| Regions | US, EU and Asia-Pacific; pinning a region costs 1.15x to 1.75x | 19 regions across the US, Europe, Asia, South America, Africa and Australia | EU only: Germany, Finland, France |
Every feature Battleships tracks (43)
| Feature | Modal | Vercel Sandbox | boat |
|---|---|---|---|
| Full VM (own kernel) | No | Yes | Yes |
| Docker inside | No | Yes | Yes |
| Nested virtualization | No | No | Yes |
| Root | Yes | Yes | Yes |
| Systemd | Yes | Not published | Yes |
| Desktop GUI | No | No | Yes |
| Browser | No | No | Yes |
| Computer-use API | No | No | Yes |
| SSH | Yes | No | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | No | Yes | Yes |
| Automatic snapshots | No | Yes | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | No | Yes | Yes |
| Memory fork | Yes | No | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | No | No |
| Wake on request | No | No | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | Yes | No | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | Yes | Yes | No |
| Secret proxy | No | Yes | Yes |
| Secret proxy for any API | No | Yes | Yes |
| Preinstalled agents | No | Yes | Yes |
| Agent harness API | No | No | Yes |
| MCP server | No | Yes | No |
| GPU | Yes | 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 | Not published | Yes | Yes |
| Usage per sandbox (API) | Not published | No | Yes |
| Audit logs | Yes | Not published | No |
| SSO | Yes | Yes | No |
| SOC 2 | Yes | Yes | No |
| HIPAA | Yes | Yes | No |
| EU data residency | Yes | Yes | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
All three ran the same hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU / 8 GB on 29 September 2026. Modal's default sandbox did 12.79 runs per second and Vercel Sandbox did 12.96, so the two are close to even. boat did 26.69, about twice the work of either in the same time.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Modal measured on: Sandbox (gVisor), request = size, any region.
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 | Modal, a month | Vercel Sandbox, 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 | $15 · Starter · with limits | $57 · Pro | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $178 · Starter · with limits | $236 · Pro · 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 | $2,229 · Starter · with limits | $2,884 · 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 | $792 · Team | $828 · Pro | $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 | $3,193 · Team · with limits | $3,152 · Pro | $2,000 · $2000/mo · with limits |
Modal: Modal bills by the second for whichever is higher, the CPU and memory you request or what you use, and an idle sandbox pays until it is terminated. Starter is free with $30 of monthly credit, Team is $250 a month with $100 of credit, and pinning a region multiplies the price by 1.15x to 1.75x.
Vercel Sandbox: Vercel charges for CPU only while your code uses it, but for memory 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.
Vercel Sandbox: boat charges one hourly rate per machine size, by the second, only while the sandbox runs; a stopped sandbox costs nothing. Plans start at $20 a month and the whole fee becomes usage credit.
Limits in those scenarios
Both cap a session at 24 hours, so the always-on agent does not fit either: $178 a month on Modal and $236 on Vercel Sandbox, against $26 on boat. On Modal the root disk resets when a sandbox ends and Docker needs the VM runtime, so the workday agent and the 50-agent fleet do not fit the default sandbox, while Vercel fits both. For 1,000 short jobs at once, Vercel fits at $3,152 a month; boat does not, because even its $2,000 plan allows 600 starts an hour and queues bursts above 90 a minute.
| Scenario | Modal | Vercel Sandbox | 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 | None that bite |
| 1,000 short jobs at once |
| None that bite |
|
Which one should you use?
Use Modal when
your sandboxes need GPUs, sit next to models you already run on Modal, or are huge bursts of short jobs.
- GPUs next to your sandboxes. Useful if your agent trains or runs models: Modal rents GPUs from T4 to B300 on the same platform.
- Huge bursts of short jobs. Modal runs up to 5,000 containers at once on Team, built for many sandboxes that finish in minutes.
- Memory snapshots. Useful for exact clones: Modal can snapshot a sandbox's memory, though this is still alpha and snapshots expire after 7 days.
Use Vercel Sandbox when
you want Docker and a real kernel in every short-lived sandbox, files that come back after a stop, or regions close to users worldwide.
- Docker and a real kernel by default. Every Vercel sandbox is a Firecracker microVM, so Docker runs with sudo and no special runtime.
- Files come back after a stop. A stopped Vercel sandbox saves a filesystem snapshot, and the next session boots from it with your files in place.
- Regions all over the world. Useful if your users are far from Europe: Vercel Sandbox runs in 19 regions, from Virginia to Tokyo and Sydney.
Where boat fits
Modal and Vercel Sandbox are both built for short runs: each stops a session at 24 hours and neither has a screen. If your agents work for hours on one machine, building, testing and checking what they built, boat gives each one a full VM that keeps its disk and is not cut off.
No 24-hour cutoff
Both Modal and Vercel Sandbox end a session at 24 hours. On Modal everything outside a Volume is lost; on Vercel the files come back but every process starts over. A boat machine has no session cap on paid plans.
A desktop and a browser built in
Neither has a desktop, so an agent cannot open the app it built and check it. On Modal you set up VNC yourself and on Vercel you build it from scratch; boat streams a 60fps desktop with Chromium and a computer-use API.
About twice the CPU work
In the same benchmark at 4 vCPU / 8 GB, boat did 26.69 runs per second, Modal 12.79 and Vercel Sandbox 12.96. Installs, builds and tests wait about twice as long on either, and you pay for the waiting.
Use boat when your agents work for hours on one machine: builds, tests, Docker, emulators and a desktop with a browser, on a fast VM that keeps its disk and is never cut off. Compare it one to one: boat vs Modal and boat vs Vercel Sandbox.
Questions people ask
Is Modal cheaper than Vercel Sandbox?
Usually a little. Sandboxes for your users cost $792 a month on Modal and $828 on Vercel, and 50 parallel agents cost $2,229 against $2,884, though that fleet does not fit Modal's default sandbox because it lacks Docker. On boat the same two cost $200 and $317.
Which is faster, Modal or Vercel Sandbox?
On CPU they are close: in the same benchmark at 4 vCPU / 8 GB, Modal did 12.79 runs per second and Vercel Sandbox 12.96. boat did 26.69, about twice either.
Can Modal and Vercel Sandbox run Docker?
Vercel Sandbox runs Docker with sudo in every sandbox, since each is a Firecracker microVM. Modal runs Docker only on its VM runtime, not the default gVisor sandbox. Every boat sandbox runs Docker with root and systemd.
How long can a sandbox run on Modal or Vercel?
Both stop a session at 24 hours. Modal's default timeout is 5 minutes and a finished sandbox is gone; Vercel's Hobby plan stops at 45 minutes, and a stopped sandbox resumes from its filesystem snapshot with processes starting over. boat has no session cap on paid plans.
Do Modal or Vercel Sandbox have GPUs?
Modal does, from T4 up to B300, on its default gVisor runtime. Vercel Sandbox and boat have no GPUs, so for GPU jobs Modal is the pick.
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