boat vs Celesto Cloud
Both give an agent its own small Linux computer with its own kernel. boat preinstalls Docker, adds /dev/kvm and seven coding agents, and runs 100 sandboxes at once from its $20 plan. Celesto gives sudo so you install Docker yourself, has no monthly fee and an open-source core, and its paid offer runs up to 10 sandboxes at once.
Where boat stands out
Docker, KVM and a bigger machine ready at start
Every boat sandbox is a full VM with root, systemd, Docker preinstalled and /dev/kvm, up to 16 vCPU / 32 GB with 251 GB of disk. Celesto gives sudo and documents a Docker install, but its docs do not cover nested virtualization, and its biggest size is 8 vCPU / 32 GB.
Room for a real fleet
boat's $20 plan runs 100 sandboxes at once, with 12 starts a minute and 200 a day included. The $2,000 plan runs 2,000. Celesto's paid offer allows up to 10 running sandboxes, subject to host capacity.
Agents, keys and hosting built in
Claude Code, Codex and five other agents come preinstalled, and you prompt, steer and follow them through one API. With the optional paid credentials proxy on, supported model keys stay outside the sandbox. Each sandbox can serve 50 token-gated HTTPS ports.
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
Celesto Cloud
Celesto AI says it gives agents real computers: a hosted API that creates an isolated Linux computer for each agent, on top of an open-source microVM core.
Celesto Cloud's Computers API creates a Linux computer from a template (scratch, coding-agent or browser-agent), runs commands in it, publishes up to four ports as HTTPS URLs, and stops, starts or deletes it. You get root or sudo inside each computer. The same open-source core, SmolVM, runs on your own machine with Firecracker on Linux or QEMU on macOS. Celesto also sells Celesto CI, which runs GitHub workflow files on its own runners with Docker preinstalled.
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 | Celesto Cloud: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A software factory runs many agents with Docker, databases and test suites at once; Celesto's paid offer stops at 10 running sandboxes, and you install Docker yourself on each one. | 5th cheapest of 365 $634 a month | 39th of 365, with limits $1,408 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Both can run a computer for as long as you like; on boat, turn off auto-stop (default 1 hour) on a paid plan. boat adds preinstalled coding agents you steer by API, and its paid credentials proxy keeps supported model keys out of the sandbox. | 6th cheapest of 361 $657 a month | 57th of 361, with limits $1,095 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Short code runs on a small machine are where Celesto does well: it has a 1 vCPU size at $0.015 an hour, no monthly fee and a no-card trial. | 189th of 336, with limits $2,000 a month | 66th of 336, with limits $50 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 needs more than 10 running at once. boat's HTTPS URLs are token-gated by default; Celesto's published URLs are public, so you add auth in your service. | 3rd cheapest of 349 $209 a month | 56th of 349, with limits $349 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 39th of 349, with limits $2,000 a month | 34th of 349, with limits $1,500 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 | 117th of 359, with limits $400 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Celesto CI runs GitHub workflow files on runners with Docker preinstalled. boat fits CI that also needs /dev/kvm for Android emulators or VMs, which Celesto does not document. | 2nd cheapest of 374 $86 a month | 148th of 374, with limits $192 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Both keep the disk across stop and start, so either works as a dev box. boat keeps a stopped disk for at least 14 days and Celesto keeps a stopped root disk free for 30 days; boat serves up to 50 HTTPS ports. | 19th cheapest of 371 $26 a month | 48th cheapest of 371 $58 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month boat streams a 60fps Linux desktop with computer tools for agents built in. Celesto's browser-agent template runs Chromium and a desktop you drive yourself over VNC and CDP. | 7th cheapest of 367 $200 a month | 73rd of 367, with limits $400 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 | 61st of 356, with limits $150 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 | 121st of 343, with limits $75 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 | 60th of 327, with limits $1,500 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month Celesto can create a computer with the internet turned off, which suits isolated attack targets; its paid offer runs up to 10 at once. | 3rd cheapest of 351 $200 a month | 35th of 351, with limits $75 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month | Not ranked: misses a requirement | Not ranked: misses a requirement |
| 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 | Celesto Cloud |
|---|---|---|
| Isolation | Full VM with its own kernel, on physical hosts dedicated to boat (AMD Ryzen 7 9700X and Ryzen 9 9950X); vCPUs are shared | MicroVM with its own kernel (Firecracker in the open-source core); vCPUs are shared |
| Biggest machine | 16 vCPU / 32 GB with 251 GB disk, on the $100 plan and above | 8 vCPU / 32 GB with 100 GB disk (XL) |
| Root and Docker | Root, systemd and Docker preinstalled in every sandbox | sudo inside every computer; you set up Docker yourself with the documented install (Standard size or larger) |
| Nested virtualization | Yes, /dev/kvm for Android emulators and VMs | Not documented for hosted computers |
| Keeping files | Disk saved every minute and on stop; stopped disk kept free for at least 14 days; writes after the last snapshot are lost if a machine fails | Root disk and workspace kept across stop and start; stopped root disks kept free for 30 days; durable CelestoFS workspace |
| Fork | Fork a running sandbox with its disk, from the last 1-minute snapshot | No fork call in the hosted Computers API or SDK |
| Sandboxes running at once | 100 on the $20 plan (12 starts a minute, 200 a day), up to 2,000 on the $2,000 plan; extra seats add 20% of those limits below the $2,000 plan | Up to 10 on the paid offer, subject to host capacity |
| Inbound access | HTTPS URLs for up to 50 ports, token-gated by default; raw ports (over IPv6 only on sandboxes without their own IPv4) | Up to four ports at a time as public HTTPS URLs; you add auth in your service |
| Outbound rules | Open internet, no provider-side allowlist | Internet on or off, set at create; no domain or IP allowlists |
| Desktop and browser | Desktop and Chrome in every sandbox, streamed at 1920x1080 and 60 fps, with computer tools for agents | browser-agent template with Chromium and a desktop; VNC display (console view or noVNC) and CDP URL; no built-in computer-use agent |
| Coding agents | Seven agents preinstalled, with a prompt, steer and events API | You set it up: the coding-agent template has Node, npm and git, and you install the agent; a Pi extension runs Pi against a remote computer |
| Keeping API keys out of the sandbox | Optional credentials proxy ($1 a day while on, plus $1 per 1,000 requests); supported model keys never enter the sandbox | No credentials proxy for hosted computers; the Pi extension keeps model keys on your own machine |
| Regions | Germany, Finland and France; no region choice, no US or Asia region; snapshot copies can be stored outside the EU | No region choice documented; AWS (United States and other regions) and Google (United States) listed as infrastructure |
| Self-hosting | Hosted service only | Open-source Apache-2.0 core runs on your own Linux or macOS machine |
| Compliance and transparency | SOC 2 is in progress. We're open about where we stand: live status and data retention policy. | No SOC 2 report published |
Every feature Battleships tracks (43)
| Feature | boat | Celesto Cloud |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | With your own setup |
| Nested virtualization | Yes | No documented way |
| Root | Yes | Yes |
| Systemd | Yes | Not published |
| Desktop GUI | Yes | Yes |
| Browser | Yes | Yes |
| Computer-use API | Yes | Yes |
| SSH | Yes | Yes |
| Persistent disk | Yes | Yes |
| Pause/resume | Yes | Yes |
| Automatic snapshots | Yes | Yes |
| Fork/clone | Yes | No documented way |
| Live fork (no pause) | Yes | No documented way |
| Memory fork | No documented way | No documented way |
| Custom image (Docker or snapshot) | Yes | Yes |
| ≥24 h sessions | Yes | Yes |
| Idle auto-stop | No documented way | No documented way |
| Wake on request | No documented way | No documented way |
| HTTPS preview URLs | Yes | Yes |
| Raw TCP inbound | Yes | Not published |
| Public IPv4 | No documented way | No documented way |
| Egress allowlist | No documented way | No documented way |
| Secret proxy | Yes | No documented way |
| Secret proxy for any API | Yes | No documented way |
| Preinstalled agents | Yes | No documented way |
| Agent harness API | Yes | No documented way |
| MCP server | No documented way | No documented way |
| GPU | No documented way | No documented way |
| Windows | No documented way | No documented way |
| Arm64 | No documented way | No documented way |
| Self-hosting | No documented way | Yes |
| BYOC | No documented way | No documented way |
| Open source | No documented way | Yes |
| Scoped API keys | Yes | Not published |
| Spend limits | Yes | Not published |
| Usage per sandbox (API) | Yes | No documented way |
| Audit logs | No documented way | Not published |
| SSO | No documented way | No documented way |
| SOC 2 | No documented way | No documented way |
| HIPAA | No documented way | No documented way |
| EU data residency | No documented way | No documented way |
| Zero data retention | Yes | Not published |
Performance
Celesto Cloud is not in the public benchmarks we use, so we cannot say which is faster. In StarSling's run of 3 October 2026 (run 37073680030), boat did 28.06 runs per second on the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, second of 15. That boat sandbox ran on an AMD Ryzen 9 9950X host; boat also places sandboxes on Ryzen 7 9700X hosts and does not promise a CPU model.
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. 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. Celesto Cloud is not in the public benchmarks we use, so we cannot say which is faster.
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 | Celesto Cloud, 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 | $14 · Usage-based paid |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $58 · Usage-based paid |
| 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 | $704 · Usage-based paid · 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 | $150 · Usage-based paid · with limits |
| 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 | $300 · Usage-based paid · with limits |
boat: Plans cost $20, $100, $500 or $2,000 a month. The whole plan price comes back as VM time for that month, and it does not carry over. A card is required, also for the 7-day free trial. You pay per second while a sandbox runs, $0.036 an hour for 4 vCPU / 8 GB. A stopped sandbox has no compute charge and keeps its disk for at least 14 days. Extra saved snapshots and the credentials proxy still charge while sandboxes are stopped.
Celesto Cloud: Celesto charges an hourly rate for each named size while a computer runs, from $0.015 an hour (Micro, 1 vCPU / 2 GB) to $0.16 (XL, 8 vCPU / 32 GB), with no monthly platform fee. The trial gives $5 of runtime credit for 30 days, Micro only, with no card. Stopped root disks are kept free for 30 days, and the egress price is not published. Rates apply to newly created sandboxes; existing sandboxes keep their original rates.
Limits in those scenarios
The limit that bites on Celesto is the 10 running sandboxes on its paid offer, which is why it does not fit 50 agents in parallel, sandboxes for your users, or 1,000 jobs at once. boat runs those fleets, but its start limits are lower than a burst needs: at 1,000 short jobs at once, even the $2,000 plan charges for starts above 90 a minute and refuses them above 270. boat never queues a start, so a refused start must be sent again. A stopped boat disk is kept for at least 14 days; keep it longer with a saved snapshot. boat has no region choice and no GPU.
| Scenario | boat | Celesto Cloud |
|---|---|---|
| One coding agent, workdays | 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 |
|
| 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 Celesto Cloud when
you want an open-source sandbox you can also run on your own machines, a handful of small sandboxes with no monthly fee, or CI runners with Docker ready.
- Open source you can run yourself. Teams that want to run sandboxes on their own hardware can use the Apache-2.0 core with Firecracker or QEMU.
- CI runners with Docker preinstalled. Celesto CI runs your GitHub workflow files on its own runners, with Docker preinstalled.
- No monthly fee, offline sandboxes and a big workspace disk. You pay only while a computer runs, a sandbox can be created with the internet off, and its workspace can grow past the root disk.
Use boat when
your agents need Docker, /dev/kvm, a desktop and a coding agent ready at start, or when you run more than 10 sandboxes at once.
- Docker, KVM and a bigger machine ready at start. Agents that run docker compose or boot an Android emulator can start at once on boat. On Celesto, Docker is a setup step on each computer, and an emulator or VM inside the computer is not documented.
- Room for a real fleet. Fifty agents in parallel, or one sandbox per user of your product, does not fit in 10 slots. On Celesto the rest wait in your own queue. On boat, a start above the plan's limit costs extra or fails with 429, and you send it again.
- Agents, keys and hosting built in. On Celesto you install and drive the agent yourself, and its keys sit inside the machine unless you run the agent from your own laptop, as the Pi extension does. Celesto publishes up to four ports as public URLs, and you add auth in your service.
Questions people ask
Is boat cheaper than Celesto?
It depends on use. A 4 vCPU sandbox that never stops costs about $26 a month on boat (8 GB) and $58 on Celesto Large (16 GB). For one coding agent on workdays, about 176 hours, Celesto Large costs about $14, less than boat's $20 plan. Below about $20 of use a month, Celesto costs less because it has no monthly fee. Fleets above 10 running sandboxes do not fit Celesto's paid offer.
Can I run Docker in a Celesto computer?
Yes. Celesto documents installing Docker with sudo on a Standard or larger computer, and its CI runners ship Docker preinstalled. boat preinstalls Docker in every sandbox and adds /dev/kvm for emulators and VMs.
How many sandboxes can I run at once?
Celesto's paid offer allows up to 10 running sandboxes, subject to host capacity, and the trial allows one. boat allows 100 on the $20 plan (12 starts a minute, 200 a day) and up to 2,000 on the $2,000 plan. Below the $2,000 plan, each extra seat adds 20% of the plan's limits.
Can I self-host Celesto?
Yes. Its core is open source under Apache-2.0 and runs on your own Linux machine with Firecracker or your Mac with QEMU. boat is only offered as a hosted service.
Where do Celesto sandboxes run?
Celesto does not let you pick a region. Its subprocessor list names AWS (United States and other regions) and Google (United States) as infrastructure. boat sandboxes run on machines in Germany, Finland and France, with no region choice, and snapshot copies can be stored outside the EU.
Is boat SOC 2 compliant?
SOC 2 is in progress; boat has no SOC 2 report today. Live status is at status.ascii.dev and the data retention policy is at docs.boat.dev/data-retention.
Start a sandbox
Install the CLI, then run boat new.
7-day free trial