boat vs Modal
boat gives each agent a full VM that saves its disk on its own, stops with no compute charge and resumes later. Modal gives you fast serverless sandboxes inside a large AI compute platform, with GPUs, regions around the world and a VM runtime for Docker.
Where boat stands out
A computer that keeps its work with no code
A boat sandbox saves its disk every minute while it runs and once more at each stop. A stopped sandbox has no compute charge. boat keeps its disk for at least 14 days, and a saved snapshot keeps it longer. Resume brings back your files and installed packages; memory, running programs and /tmp do not come back. Modal sandboxes have no paused state and last at most 24 hours. To keep work, your code takes a filesystem snapshot or writes to a Volume and boots a new sandbox from it.
Desktop, browser and agents ready to go
Every boat sandbox has a 1920x1080, 60 fps desktop, Chrome and computer tools for agents. Seven coding-agent harnesses are installed, and you drive them from outside through one API. An optional paid credentials proxy keeps supported providers' API keys out of the sandbox. On Modal you set these up yourself: its official example installs Xvfb, Chromium, x11vnc and noVNC in a VM sandbox and runs a Browser Use agent.
More CPU work per hour in one public run
StarSling ran the V8 Web Tooling Benchmark at 4 vCPU / 8 GB on 3 October 2026 (run 37073680030). boat did 28.06 runs/s, second of 15. Modal's VM runtime did 15.48 and its default gVisor runtime 12.60. The boat sandbox in this run ran on an AMD Ryzen 9 9950X host. The fleet also has Ryzen 7 9700X hosts, and boat does not promise a CPU model.
The companies and products
boat
boat.dev runs sandboxes for AI agents: each one is a full Linux VM with its own kernel.
You start a machine with one command or one API call, then SSH in, run Docker, open a desktop, or let a built-in agent (Claude Code, Codex, OpenCode and others) work in it. The disk stays between sessions, and a stopped sandbox costs nothing.
- Isolation
- Full VM per sandbox, own kernel
- Regions
- EU: Germany, Finland, France
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). Your code can save a sandbox's disk as a filesystem snapshot and start new sandboxes from it. It suits teams that already run models on Modal and want scalable code execution next to their GPUs.
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. “5th cheapest of 365” means 4 of the other 364 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 | boat: price rank | Modal: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Software factories need Docker, a desktop and a disk that survives between sessions. boat has all three built in. On Modal you set them up: pick the VM runtime, install the desktop yourself and snapshot the filesystem between sessions. The monthly bills in this preset are close: $633 on Modal VM against $634 on boat. | 5th cheapest of 365 $634 a month | 106th of 365, with limits $633 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An always-on agent hits Modal's 24-hour sandbox limit, so your code snapshots the filesystem and starts a new sandbox at least once a day. On boat, a paid plan lets you turn auto-stop off (default 1 hour), and then there is no session cap. | 6th cheapest of 361 $657 a month | 61st of 361, with limits $1,574 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month For huge numbers of short code runs, Modal fits better. It publishes no start-rate cap and allows 5,000 containers at once on Team. boat's $20 plan includes 12 starts a minute and 200 a day, and its $2,000 plan 90 a minute and 2,400 a day. | 189th of 336, with limits $2,000 a month | 15th cheapest of 336 $140 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Giving each of your users a sandbox that they come back to works well on boat. A stopped sandbox has no compute charge, and boat keeps its disk for at least 14 days, or longer with a saved snapshot. On Modal your code saves and restores each user's filesystem snapshot. | 3rd cheapest of 349 $209 a month | 18th cheapest of 349 $671 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month RL rollouts launch many short sandboxes at once, which is what Modal is built for. On boat, each rollout counts as a start against the plan's start limits. | 39th of 349, with limits $2,000 a month | 6th cheapest of 349 $11,982 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 10th cheapest of 359 $200 a month | 62nd cheapest of 359 $889 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 in every boat sandbox. On Modal you set runtime="vm" to run Docker. | 2nd cheapest of 374 $86 a month | 110th cheapest of 374 $1,112 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB | 19th cheapest of 371 $26 a month | 129th of 371, with limits $9.07 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month boat ships a 60 fps desktop, Chrome and computer tools for agents. On Modal you set it up: its official example installs Xvfb, Chromium and noVNC in a VM sandbox. With that setup, this preset costs $153 a month on Modal against $200 on boat. | 7th cheapest of 367 $200 a month | 45th cheapest of 367 $153 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month | 7th cheapest of 356 $200 a month | 57th of 356, with limits $121 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 6th cheapest of 343 $90 a month | 98th cheapest of 343 $252 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 73rd of 327, with limits $1,800 a month | 171st of 327, with limits $5,957 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month | 3rd cheapest of 351 $200 a month | 289th of 351, with limits $464 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 boat has none, so Modal is the pick for GPU jobs. | Not ranked: misses a requirement | 18th cheapest of 102 $1,116 a month |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Not ranked: misses a requirement | Not ranked: misses a requirement |
Key features
| Feature | boat | Modal |
|---|---|---|
| Isolation | Full VM with its own kernel, every sandbox | gVisor by default; VM runtime available (Modal VM Sandboxes) |
| Docker inside | Yes, in every sandbox | Yes with the VM runtime (runtime="vm"); not documented on the default gVisor runtime |
| Nested virtualization | Yes (/dev/kvm, Android emulator) | VM runtime only, Enterprise plan, as an experimental option |
| Desktop, browser, computer use | 1920x1080, 60 fps desktop, Chrome and computer tools built in | No built-in desktop; Modal's official example installs Xvfb, Chromium, x11vnc and noVNC in a VM sandbox and runs a Browser Use agent |
| Stop and resume | Stop with no compute charge; resume with your files and installed packages, with no code. Memory, running programs and /tmp are not kept | No paused state; your code saves the root disk with a filesystem snapshot or a Volume, then starts a new sandbox from it |
| Saved state retention | Stopped disk kept at least 14 days; saved snapshots have no expiry (10 per $20 of plan included, then $0.056 a day each) | Filesystem snapshots kept 30 days by default, or indefinitely; storage price not published |
| Longest session | No cap once auto-stop is off on a paid plan (default 1 h, timer up to 30 days) | 24 hours (default 5 minutes); for longer work Modal recommends a filesystem snapshot and a new sandbox |
| Fork | Fork a running sandbox without stopping it, from its last 1-minute snapshot | One filesystem snapshot boots many identical sandboxes |
| Memory snapshots | No (a resume is like a reboot) | Yes, in alpha: snapshotting ends the sandbox, snapshots expire after 7 days |
| Disk | 51 GB for your files at 4 vCPU; more disk needs a bigger size (up to 251 GB) | 512 GiB by default |
| Custom images | Start from boat's image or a saved snapshot; you cannot boot your own Docker image | Build images in code or pull any linux/amd64 registry image |
| Start limits | $20 plan: 12 a minute, 200 a day included; $2,000 plan: 90 a minute, 2,400 a day. Paid extra starts up to 6x, then refused | No start-rate cap published; 100 containers at once on Starter, 5,000 on Team |
| GPUs | None | T4 to B300; 10 at once on Starter, 50 on Team |
| Outbound network rules | No provider-side allowlist | Block all, IP-range allowlist, domain allowlist (beta) |
| Regions | Germany, Finland and France; no US or Asia region, no region choice | US, EU and Asia-Pacific plus narrower regions; pinning costs 1.15x to 1.75x |
| Coding agents and key proxy | Seven agent harnesses preinstalled, with an API to prompt them. Optional AI credentials proxy ($1 a day + $1 per 1,000 requests) keeps supported providers' keys out of the sandbox | You set it up: install a harness in the image (official OpenCode example). Secrets are environment variables, or you run your own proxy in a sidecar (beta) |
| Hosting ports | HTTPS URLs for up to 50 ports, token-gated by default; raw ports (IPv6 only on sandboxes without their own IPv4) | Encrypted and raw TCP tunnels, connect tokens, inbound IP allowlist |
| Compliance | SOC 2 in progress; no HIPAA BAA documented | SOC 2 Type II; HIPAA on the Enterprise plan |
| Compliance and transparency | SOC 2 is in progress. We're open about where we stand: live status and data retention policy. | SOC 2 |
Every feature Battleships tracks (43)
| Feature | boat | Modal |
|---|---|---|
| Full VM (own kernel) | Yes | No documented way |
| Docker inside | Yes | No documented way |
| Nested virtualization | Yes | No documented way |
| Root | Yes | Yes |
| Systemd | Yes | No documented way |
| Desktop GUI | Yes | With your own setup |
| Browser | Yes | With your own setup |
| Computer-use API | Yes | With your own setup |
| SSH | Yes | With your own setup |
| Persistent disk | Yes | With your own setup |
| Pause/resume | Yes | No documented way |
| Automatic snapshots | Yes | No documented way |
| Fork/clone | Yes | Yes |
| Live fork (no pause) | Yes | Not published |
| Memory fork | No documented way | Yes |
| Custom image (Docker or snapshot) | Yes | Yes |
| ≥24 h sessions | Yes | Yes |
| Idle auto-stop | No documented way | Yes |
| Wake on request | No documented way | No documented way |
| HTTPS preview URLs | Yes | Yes |
| Raw TCP inbound | Yes | Yes |
| Public IPv4 | No documented way | No documented way |
| Egress allowlist | No documented way | Yes |
| Secret proxy | Yes | No documented way |
| Secret proxy for any API | Yes | No documented way |
| Preinstalled agents | Yes | With your own setup |
| Agent harness API | Yes | No documented way |
| MCP server | No documented way | No documented way |
| GPU | No documented way | Yes |
| Windows | No documented way | No documented way |
| Arm64 | No documented way | No documented way |
| Self-hosting | No documented way | No documented way |
| BYOC | No documented way | No documented way |
| Open source | No documented way | No documented way |
| Scoped API keys | Yes | Not published |
| Spend limits | Yes | Not published |
| Usage per sandbox (API) | Yes | Not published |
| Audit logs | No documented way | Yes |
| SSO | No documented way | Yes |
| SOC 2 | No documented way | Yes |
| HIPAA | No documented way | Yes |
| EU data residency | No documented way | Yes |
| Zero data retention | Yes | Not published |
Performance
StarSling's run of 3 October 2026 (run 37073680030) ran the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, median of 5 sandboxes. boat did 28.06 runs/s, Modal's VM runtime 15.48 and its default gVisor runtime 12.60. StarSling ran Modal with cpu=4, and the sandbox saw 4 CPUs (PR #121). The boat sandbox ran on an AMD Ryzen 9 9950X host; other boat hosts are Ryzen 7 9700X. Providers ran on different hardware in different regions.
V8 Web Tooling Benchmark at a 4 vCPU / 8 GB target, runs per second, higher is better. Source: StarSling's run of 3 October 2026 (run 37073680030), median of 5 sandboxes. Providers ran on different hardware in different regions. StarSling gives a shared rank when the run cannot separate two providers. Modal measured on: Sandbox (gVisor), request = size, any region. Every boat sandbox in this run ran on an AMD Ryzen 9 9950X host. boat also places some sandboxes on AMD Ryzen 7 9700X hosts, which this run did not measure.
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. Start limits, per-start fees and plan minimums are in the price, boat's included.
| Scenario | boat, a month | Modal, 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 | $20 · $20/mo | $0 · Starter |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $109 · Starter · with limits |
| 50 agents in parallel 50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside | $317 · $20/mo × 3 seats | $1,480 · Starter · with limits |
| Sandboxes for your users 2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once | $200 · $100/mo × 2 seats | $579 · Team |
| 1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month | $8,800 · $2000/mo · with limits | $1,911 · Team · with limits |
boat: boat charges one hourly rate per size, by the second, while the sandbox runs: $0.036 an hour for 4 vCPU / 8 GB. Billing starts when the machine is provisioned. Plans cost $20, $100, $500 or $2,000 a month, and the whole plan price comes back as VM time for that month. Included VM time does not carry over. A card is required, also for the 7-day free trial.
boat: A stopped sandbox has no compute charge. Saved snapshots above the plan's included count ($0.056 a day each) and the optional AI credentials proxy still charge while sandboxes are stopped. Create, fork and resume each count as a start. Above the included starts, an extra start costs $0.03 to $0.07, and above 6 times the included number a start fails with 429.
Modal: Modal bills by the second for whichever is higher, the CPU and memory you request or what you use. Sandboxes bill at 3x Modal's list rate ($0.0000131 per core-second and $0.00000222 per GiB-second list). An idle sandbox pays until it is terminated. Starter is free with $30 of monthly credit, and Team is $250 a month with $100 of credit. Pinning a region multiplies the price by 1.15x to 1.75x, and egress above the free allowance costs $0.04 per GiB.
Modal: How we count Modal's CPU: Modal's pricing page lists CPU per "Physical core (2 vCPU equivalent)". We use that conversion, the lowest Modal price we can defend. This page prices a 4 vCPU / 8 GiB sandbox as 2 cores: $0.476 an hour at a full request. Know this before you size: with cpu=2, the sandbox shows 2 CPUs, not 4. StarSling ran Modal with cpu=4, and the sandbox saw 4 CPUs (PR #121). If your code needs 4 CPUs, Modal bills 4 cores: $0.76 an hour. The scenario bills use the cheapest Modal setup that Battleships prices for each workload. For some, that is a small request billed close to actual use, with no guaranteed CPU above the request.
Limits in those scenarios
On Modal, a sandbox lives at most 24 hours. Its root disk ends with it, so the workday, always-on and fleet scenarios keep only files your code saves to a filesystem snapshot or a Volume. Docker needs the VM runtime, so the 50-agent fleet does not fit the default gVisor sandbox. On boat, the limit that bites is starts. The $20 plan includes 12 starts a minute and 200 a day; the $2,000 plan includes 90 a minute and 2,400 a day. Create, fork and resume each count as a start. For 1,000 jobs at once, boat pays for extra starts and still refuses starts above 270 a minute, so that scenario does not fit. boat also has no GPU and no region outside Germany, Finland and France.
| Scenario | boat | Modal |
|---|---|---|
| One coding agent, workdays | None for this scenario | None for this scenario |
| One agent that never stops | None for this scenario |
|
| 50 agents in parallel | None for this scenario |
|
| Sandboxes for your users | None for this scenario | None for this scenario |
| 1,000 short jobs at once |
|
|
boat's own limits
- Plans start at $20 a month. The plan price comes back as VM time for that month. Unused time does not carry over. Source
- Starts (create, fork, resume): the $20 plan includes 12 a minute, 60 an hour and 200 a day. The $2,000 plan includes 90 a minute, 600 an hour and 2,400 a day. Extra starts cost $0.03 to $0.07. A start fails with 429 above 3 times the included starts per minute, or 6 times per hour or per day. Source
- A stopped sandbox has no compute charge. boat keeps its disk for at least 14 days. A saved snapshot keeps it longer; extra saved snapshots and the AI credentials proxy are paid. Source
- Auto-stop defaults to 1 hour. You can turn it off on a paid plan. The trial allows 2 hours at most. Source
- boat saves the disk every minute and at each stop. Memory, running programs and /tmp are not saved: a resume is like a reboot. A machine failure loses writes after the last good snapshot. Source
- Machines run in Germany, Finland and France only. Snapshot copies can be stored outside the EU. Source
- Public IPv6 and outbound IPv4. No dedicated IPv4. The IP changes at each resume. Source
- Hosts are AMD Ryzen 7 9700X and Ryzen 9 9950X. vCPUs are shared. No GPU. You cannot boot your own Docker image. Source
Which one should you use?
Use Modal when
your sandboxes need GPUs, must run in the US or Asia, need an outbound allowlist, or come in huge bursts of short jobs. It also fits when your team already runs models on Modal.
- GPUs next to your sandboxes. Useful if your agent trains or runs models: Modal rents GPUs from T4 to B300 on the same platform as its sandboxes.
- More regions and outbound rules. Modal runs in the US, EU and Asia-Pacific, and can block a sandbox's outbound traffic or allow only chosen domains.
- Very large fleets and memory snapshots. Useful for huge bursts of short jobs: Modal runs up to 5,000 containers on Team and publishes no start-rate cap.
- Bigger disk and custom images. Modal gives a sandbox 512 GiB of disk by default and boots any linux/amd64 registry image.
Use boat when
your agents work for hours or days and need Docker, a desktop and a disk kept between sessions with no code. Turn auto-stop off on a paid plan for long work (default 1 hour).
- A computer that keeps its work with no code. On Modal you write the snapshot-and-restore step yourself. On boat every stop keeps the disk with no code.
- Desktop, browser and agents ready to go. An agent that sees a screen can open the app it built and check it. On boat that works from the first start. On Modal you build and keep up the image for it, and secrets are environment variables unless you run your own proxy.
- More CPU work per hour in one public run. In that run, installs, builds and tests ran about 1.8x faster than on Modal's VM runtime and 2.2x faster than on gVisor. On the Better-Auth repo boat took 135.8 s, Modal VM 180.8 s and Modal gVisor 326.9 s. Providers ran on different hardware in different regions, and boat did not lead every test: it ranked 9th of 15 on SQLite Speedtest.
Questions people ask
Is boat cheaper than Modal?
It depends on the workload. One always-on agent costs $26 a month on boat against $109 on Modal, but Modal stops each sandbox after 24 hours. 50 parallel agents cost $317 against $1,480 on Modal's VM runtime, and 30,000 short user sessions cost $200 against $579. For one agent on workdays Modal is cheaper: $0 inside its $30 monthly credit, against $20 on boat, and your code keeps files between days with a filesystem snapshot. For 1,000 short jobs at once, Modal costs $1,911 a month, and boat costs $8,800 and still hits its start limits.
Can Modal sandboxes run Docker?
Yes, with the VM runtime, which you choose with runtime="vm". Modal tells you to start dockerd as the sandbox entrypoint. The default gVisor runtime does not document Docker support. Every boat sandbox runs Docker on its own kernel with no extra setting.
How long can a Modal sandbox run?
The default timeout is 5 minutes and the maximum is 24 hours. For longer work, Modal recommends taking a filesystem snapshot and starting a new sandbox from it. On boat, auto-stop defaults to 1 hour; on a paid plan you can turn it off, and then there is no session cap.
Can a Modal sandbox keep its files between sessions?
Yes, if your code does it. A filesystem snapshot saves the whole root disk, installed packages included, and is kept 30 days by default or indefinitely. A Volume keeps one folder. boat saves the disk every minute and at each stop with no code, and keeps it at least 14 days, or longer with a saved snapshot.
Can a Modal sandbox run a desktop and a browser?
Yes, with your own setup. Modal's official example installs Xvfb, Chromium, x11vnc and noVNC in a VM sandbox and runs a Browser Use agent. boat has a 60 fps desktop, Chrome and computer tools built into every sandbox.
What is the difference between Modal Sandboxes and Modal VM Sandboxes?
Both use the same API and prices. gVisor is the default runtime and can use GPUs. Set runtime="vm" for a real Linux kernel with Docker and FUSE, but no GPUs. Battleships compares them as two separate cards.
Does boat have GPUs?
No. boat sandboxes are CPU machines, and graphics run on the CPU. If your workload needs a GPU in the sandbox, Modal is the better fit.
Is boat SOC 2 compliant?
SOC 2 is in progress, and boat has no SOC 2 report today. Modal has SOC 2 Type II. We're open about where we stand: live status at status.ascii.dev and our data retention policy at docs.boat.dev/data-retention.
Start a sandbox
Install the CLI, then run boat new.
7-day free trial
More comparisons
- boat vs E2B
- boat vs Daytona
- boat vs Vercel Sandbox
- boat vs Blaxel
- boat vs Runloop
- boat vs Tensorlake
- boat vs Cloudflare Sandbox
- boat vs Agent 37
- boat vs exe.dev
- boat vs Instabox
- boat vs Oracle Cloud
- boat vs IONOS Cloud
- E2B vs Modal
- Daytona vs Modal
- Modal vs Vercel Sandbox
- Modal vs Cloudflare Sandbox
- Modal vs Blaxel
- Modal vs Runloop
- Modal vs Tensorlake
- Modal vs Morph Cloud