boat vs Amika
Amika is a layer for coding-agent workstations: it puts rigs on sandbox clouds such as E2B, Daytona and Vercel, or on machines you already own. boat runs every sandbox as a full VM on its own hosts, with Docker, a desktop and a credentials proxy built in.
Where boat stands out
Real keys never enter the sandbox
boat's AI credentials proxy adds your model keys, and keys for any HTTPS API you register, on the way out, so the sandbox only ever sees a stand-in. When an account runs out of credits, boat can switch the credential to another account of the same type. On Amika, secrets are copied into the rig as environment variables, and a proxy for them is still an early design on its roadmap.
One full VM, the same everywhere
Every boat sandbox is a full VM on boat's own AMD Ryzen 9 9950X hosts, with root, systemd, Docker and /dev/kvm ready. Amika places hosted rigs on other sandbox clouds such as E2B, Daytona and Vercel, so what you get underneath depends on the provider. Docker needs a separate preset, and /dev/kvm is not documented.
A desktop and a browser in every sandbox
boat streams a 60fps Linux desktop with Chromium, and agents can use it through computer use and browser control. You can embed the stream in your own app. Amika rigs are headless: you reach them by SSH, an editor or a web console for the agent.
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
Amika
Amika is a small YC startup that turns cloud sandboxes and your own computers into shared devboxes, called rigs, for coding agents and the people working with them.
A rig is a VM with your repos checked out and Claude Code, Codex, OpenCode and Pi preinstalled; you reach it over SSH, an editor, or a web console, and start agents from the API, CLI, Slack or Linear. Hosted rigs run on sandbox clouds such as E2B, Daytona and Vercel, and you can also add your own machines or cloud accounts, which are never metered. It is built for teams that want many coding agents working on their repos, with shared sessions and preview links.
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. “3rd cheapest of 367” means 2 of the other 366 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 | Amika: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Amika is built for this kind of work, but a rig needs the Docker preset to run containers and there is no documented /dev/kvm; boat has Docker, systemd and /dev/kvm on every machine. | 3rd cheapest of 367 $634 a month | 334th of 367, with limits $8,777 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Amika keeps rigs running and lets you start agents from Slack or Linear; boat also keeps the disk safe with snapshots every minute and lets a stopped sandbox cost nothing. | 5th cheapest of 366 $657 a month | 344th of 366, with limits $13,806 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Amika places a little better here because it bills hosted compute per second with no monthly fee on its Free plan, though neither is a full fit. | 186th of 345, with limits $2,000 a month | 165th of 345, with limits $788 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Amika is a workstation tool for your own team, with per-user seats, so serving hundreds of sandboxes to your users is a compromise there. | Cheapest of 361 $209 a month | 175th of 361, with limits $4,271 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 34th of 360, with limits $2,000 a month | 178th of 360, with limits $19,552 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 10th cheapest of 364 $200 a month | 205th of 364, with limits $3,462 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Amika has a GitHub Actions and CI feature on its Team plan; boat runs CI jobs as full VMs with Docker and up to 8 vCPU from the $20 plan. | 4th cheapest of 377 $86 a month | 199th of 377, with limits $1,907 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Both fit a dev box that stays on; on Amika it can also be a machine you already own, which is never metered. | 18th cheapest of 368 $26 a month | 132nd of 368, with limits $562 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Amika rigs are headless, with no desktop or browser to drive; boat streams a 60fps desktop with Chromium and computer use. | 6th cheapest of 369 $200 a month | 322nd of 369, with limits $2,680 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month Amika documents no browser inside a rig, so you would install and run one yourself; boat ships Chromium with browser control. | 7th cheapest of 367 $200 a month | 326th of 367, with limits $1,410 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 6th cheapest of 346 $90 a month | 312th of 346, with limits $970 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 71st of 338, with limits $1,800 a month | 283rd of 338, with limits $19,209 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month | 2nd cheapest of 360 $200 a month | 328th of 360, with limits $1,529 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 |
| 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 |
Key features
Every feature Battleships tracks (43)
| Feature | boat | Amika |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | Yes |
| Nested virtualization | Yes | No |
| Root | Yes | Not published |
| Systemd | Yes | Not published |
| Desktop GUI | Yes | No |
| Browser | Yes | No |
| Computer-use API | Yes | No |
| SSH | Yes | Yes |
| Persistent disk | Yes | Yes |
| Pause/resume | Yes | Yes |
| Automatic snapshots | Yes | Yes |
| Fork/clone | Yes | Yes |
| Live fork (no pause) | Yes | No |
| Memory fork | No | No |
| Custom image (Docker or snapshot) | Yes | Yes |
| ≥24 h sessions | Yes | Yes |
| Idle auto-stop | No | Yes |
| Wake on request | No | No |
| HTTPS preview URLs | Yes | Yes |
| Raw TCP inbound | Yes | Not published |
| Public IPv4 | No | No |
| Egress allowlist | No | Yes |
| Secret proxy | Yes | No |
| Secret proxy for any API | Yes | No |
| Preinstalled agents | Yes | Yes |
| Agent harness API | Yes | Yes |
| MCP server | No | No |
| GPU | No | Yes |
| Windows | No | No |
| Arm64 | No | No |
| Self-hosting | No | Yes |
| BYOC | No | Yes |
| Open source | No | Yes |
| Scoped API keys | Yes | Not published |
| Spend limits | Yes | Not published |
| Usage per sandbox (API) | Yes | No |
| Audit logs | No | Yes |
| SSO | No | Yes |
| SOC 2 | No | No |
| HIPAA | No | No |
| EU data residency | Yes | No |
| Zero data retention | Yes | Not published |
Performance
There is no comparable measurement for Amika. Its hosted rigs run on other sandbox clouds, so speed depends on which provider is behind a rig. boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB in StarSling's public run of 2026-09-29.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Amika 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 | boat, a month | Amika, 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 | $122 · Free · with limits |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $562 · Free · 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 | $5,758 · Team (50 concurrent) · 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 | $2,228 · Team (300 concurrent) · with limits |
| 1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month | $2,000 · $2000/mo · with limits | $5,518 · Team (1,000 concurrent) · with limits |
boat: You pick a monthly plan from $20, and all of it becomes usage credit spent per second on running sandboxes. A stopped sandbox costs nothing and keeps its disk.
Amika: Hosted rigs bill per second for the vCPU, memory and storage they are given: $0.1008 per vCPU-hour, $0.0324 per GiB-hour of memory and $0.00216 per GiB-hour of storage, with a one-time $50 of credit to start. The Team plan adds $30 per active user a month, plus $2 a month for each sandbox at once past 10; your own machines and AI tokens are never billed.
Limits in those scenarios
Amika's Team plan includes 10 sandboxes at once started by API keys, then charges $2 a month for each one past that, so the table prices 50, 300 and 1,000 at once as that fee rather than a wall. Sandboxes people start by hand are never counted. Hosted rigs are still in beta, with prices that may change, so no scenario counts as a full fit on Amika. On boat the limit is a sandbox count (100 to 2,000 by plan) plus starts per minute, so the 1,000-job burst gets queued and needs more starts per day than the $2,000 plan allows.
| Scenario | boat | Amika |
|---|---|---|
| One coding agent, workdays | None that bite |
|
| One agent that never stops | None that bite |
|
| 50 agents in parallel | None that bite |
|
| Sandboxes for your users | None that bite |
|
| 1,000 short jobs at once |
|
|
Which one should you use?
Use Amika when
your team wants coding agents on machines it already owns, started from Slack or Linear, with an open source core.
- Run agents on machines you own. Useful if you already have spare hardware or cloud credits: Amika turns them into rigs and never meters them.
- Built around team workflows. Useful if your team starts agents from Slack or Linear and wants to share their sessions and preview links.
- Open source core and GPUs. Useful if you want to read the code that runs in your VMs, or need GPU machines for agents.
Use boat when
your agents need a full VM that is the same every time, with Docker, /dev/kvm, a desktop and a browser, and keys that never enter the sandbox.
- Real keys never enter the sandbox. A coding agent that reads environment variables can print or leak a key it was given. On Amika the key sits inside the rig, so a prompt injection in a repo can reach it.
- One full VM, the same everywhere. Builds that use Docker, emulators that need /dev/kvm, or services run by systemd can work on one Amika rig and fail on another. On boat every machine has all of them.
- A desktop and a browser in every sandbox. An agent that must check a web page it built, click through an app or log in to a site needs a screen. On Amika you would install and run a browser and a display yourself in each rig.
Questions people ask
Is boat cheaper than Amika?
In the scenarios we priced, yes: one coding agent on workdays comes to $20 a month on boat and $122 on Amika, and one agent running all month comes to $26 versus $562. 50 agents in parallel come to $317 versus $5,758. If you run Amika only on machines you already own, its compute is not metered, so you pay only for those machines and seats.
Does Amika run its own servers?
Amika's hosted rigs run on other sandbox clouds such as E2B, Daytona and Vercel, and you can also connect your own machines and cloud accounts. boat runs every sandbox on its own dedicated AMD Ryzen 9 9950X hosts in the EU.
Is Amika open source?
Partly. The CLI, the daemon inside each rig, the rig images and the TypeScript SDK are open source under Apache-2.0. The control plane, the web UI and the relay are closed source. boat is closed source.
Can agents on Amika use a browser or desktop?
Amika documents no desktop, browser or computer use inside a rig; you work in it over SSH, an editor or the OpenCode web console. boat ships a 60fps desktop with Chromium, computer use and browser control in every sandbox.
Where does my data run?
boat runs only in the EU (Germany, Finland and France), with EU data residency and a zero data retention option. Amika does not publish a region for hosted rigs, and its privacy policy says data may be processed in the United States and other countries, unless you run rigs on your own machines.
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.
Start a sandbox
Install the CLI, then run boat new.
7-day free trial