Agent 37 vs Instabox
Agent 37 starts each agent in a gVisor sandbox that sleeps with its memory kept and wakes on the next request. It sells a full VM type at 2.5x its default compute rate and bills per minute from a wallet. Instabox gives every machine its own Firecracker microVM with root and Docker, pause with memory, a free plan for one machine and paid plans from $20 a month.
The companies and products
Agent 37
Agent 37 hosts AI agents, one isolated, always-on sandbox per user, behind one API.
You create an instance from a template (Hermes, OpenClaw, Claude Code, Codex, OpenCode, Pi, Grok or your own Docker image) and talk to it through a chat API, files, exec, SSH and HTTPS URLs. The default instance is a container enforced by gVisor. A VM type gives a full Linux machine with its own kernel for Docker, Kubernetes and Android emulators. It is built for companies that give each of their customers an agent.
Instabox
Instabox gives each AI agent its own Linux machine in the cloud. It has a free plan with one machine and paid plans from $20 a month plus usage.
Each Instabox machine is a Firecracker microVM with its own kernel, root and Docker. It starts from a template (agent-browser by default, universal, t3, herdr, orca, android, omarchy or ci), from your own snapshot, or from a Docker image that you import as a snapshot. The default size is 4 vCPU / 8 GiB, and paid plans go up to 32 vCPU / 64 GiB. You drive it from a CLI, a TypeScript SDK, an HTTP API or an MCP server, with SSH, mosh, HTTPS port URLs and a desktop for computer use.
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. “4th cheapest of 365” means 3 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 | Agent 37: price rank | Instabox: price rank | boat: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Docker runs on every Instabox machine and on Agent 37's VM type, which bills compute at 2.5x. All three fork a running machine's disk. Instabox goes up to 32 vCPU / 64 GiB, boat to 16 vCPU / 32 GB and Agent 37 to 4 vCPU. | 4th cheapest of 365 $551 a month | Cheapest of 365 $410 a month | 5th cheapest of 365 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Both can keep an idle assistant with its memory. Agent 37 sleeps after a quiet period and wakes on the next request. Instabox pauses after 30 idle minutes and wakes on a port visit. | Cheapest of 361 $279 a month | 4th cheapest of 361 $534 a month | 6th cheapest of 361 $657 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month | Cheapest of 336 $22 a month | 4th cheapest of 336 $56 a month | 189th of 336, 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 Agent 37 runs up to 1,000 awake instances after $1,000 of top-ups; sleeping ones do not count. Instabox runs 100 machines at once on Pro, 200 on Builder and 500 on Team. boat runs 100 to 2,000 by plan, with start limits per plan. | Cheapest of 349 $79 a month | 10th cheapest of 349 $500 a month | 3rd cheapest of 349 $209 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 9th of 349, with limits $363 a month | 16th of 349, with limits $845 a month | 39th of 349, with limits $2,000 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month Instabox runs 100 to 500 machines at once by plan, at 30 to 100 starts a minute. Agent 37 runs up to 1,000 awake instances. boat runs 100 to 2,000 at once; the $20 plan includes 12 starts a minute and 200 a day. | Cheapest of 359 $58 a month | 34th cheapest of 359 $500 a month | 10th cheapest of 359 $200 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Instabox runs Docker on every machine and has a ci template for GitHub Actions. Agent 37 needs its VM type for Docker. Jobs that need /dev/kvm get it on boat and on Agent 37's VM type; Instabox does not document /dev/kvm. | 227th of 374, with limits $96 a month | 3rd cheapest of 374 $89 a month | 2nd cheapest of 374 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB All three keep a machine and its disk. Instabox adds mosh and paste-anywhere SSH tokens. Agent 37 costs the least to keep running in this page's estimate. boat adds systemd and /dev/kvm. | 3rd cheapest of 371 $12 a month | 14th cheapest of 371 $24 a month | 19th cheapest of 371 $26 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Agent 37 adds a noVNC desktop through a template build. Instabox has a VNC desktop on most templates, with a screenshot and click API. boat streams a 60fps desktop in every sandbox. | 3rd cheapest of 367 $120 a month | 4th cheapest of 367 $122 a month | 7th cheapest of 367 $200 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month | 47th of 356, with limits $29 a month | 49th of 356, with limits $61 a month | 7th cheapest of 356 $200 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | Cheapest of 343 $32 a month | 5th cheapest of 343 $83 a month | 6th cheapest of 343 $90 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 39th of 327, with limits $648 a month | 31st of 327, with limits $1,658 a month | 73rd of 327, with limits $1,800 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month All three fork a running machine's disk, so one attempt can branch into many. None copies memory, and each fork boots fresh. A boat fork uses the last 1-minute snapshot. | Cheapest of 351 $43 a month | 2nd cheapest of 351 $100 a month | 3rd cheapest of 351 $200 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 | 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 | Not ranked: misses a requirement |
Key features
| Feature | Agent 37 | Instabox | boat |
|---|---|---|---|
| Isolation | gVisor sandbox by default; full VM with its own kernel as an opt-in type | Firecracker microVM with its own kernel and root for every machine | Full VM with its own kernel, on AMD Ryzen 7 9700X and Ryzen 9 9950X hosts dedicated to boat; vCPUs are shared |
| Docker and systemd | Docker only on the VM type, at 2.5x the default compute rate; no init system, also on the VM type | Docker and Docker Compose with root and systemd on every machine | Docker, Compose and systemd in every sandbox |
| Nested virtualization | /dev/kvm on the VM type, with a ready Android template | /dev/kvm not documented; an android template runs Android 14 natively, with no emulator | /dev/kvm in every sandbox, Android emulator works |
| Machine sizes | Up to 4 vCPU on every type | 4 vCPU / 8 GiB on Free; up to 8 / 16 on Pro, 16 / 32 on Builder, 32 / 64 on Team | 2 to 8 vCPU on the $20 plan; up to 16 vCPU / 32 GB on $100 and above |
| Machines at once | 3 awake before a top-up, then 10, 50, 200 and 1,000 as top-ups grow; sleeping and stopped ones do not count | 1 on Free; 100 running on Pro, 200 on Builder, 500 on Team | 100 to 2,000 sandboxes at once by plan, with included starts per minute, hour and day |
| Sleep and pause | Opt-in auto-sleep after a quiet period, memory checkpointed, wakes on the next request | Pause keeps memory and running programs; auto-pause after 30 idle minutes; a port visit wakes it | Stop keeps the disk for at least 14 days with no compute charge; memory is not saved, so a resume is like a reboot |
| Fork and snapshots | Fork a running, stopped or sleeping instance (disk copy, the fork boots fresh); nightly backups in 7 slots plus one on-demand, restored in place | Fork a running or stopped machine, disk only, up to 5 copies per call; snapshots saved by hand | Fork a running sandbox from its last 1-minute snapshot; disk saved every minute and at stop; saved snapshots that new sandboxes start from |
| Custom images | Your own linux/amd64 Docker image as a template | Import a Docker or OCI image as a snapshot, then start machines from it | Start from boat's image or a saved snapshot; you cannot boot your own Docker image |
| Coding agents | Hermes, OpenClaw, Claude Code, Codex, Grok, OpenCode, Pi, with a chat API per instance | Claude Code, Codex and OpenCode on some templates; CLI, skills and MCP; you run the agent yourself | Seven agents installed, driven from outside with a prompt, steer and events API |
| Keeping API keys out of the sandbox | Managed LLM, search and Composio through a per-instance token; your own keys go in as env vars | Secrets proxy swaps a placeholder for the real key in outbound HTTPS headers to allowed hosts | Optional paid proxy ($1 a day plus $1 per 1,000 requests) for model providers and HTTPS APIs you register; other secrets still enter the sandbox |
| Desktop and computer use | You set it up: noVNC desktop through a template build, or the Ubuntu VM template | VNC desktop with a browser on most templates, screenshot and click API, shareable view or control links | 60fps desktop with Chrome in every sandbox, with computer tools for the built-in agents |
| Port URLs | HTTPS URL per port, signed URLs, custom domains; no public IP | HTTPS on up to 16 ports, public by default, token-protected with --private; raw TCP through the port gateway | HTTPS on up to 50 ports, token-protected by default; raw ports over IPv6 only on sandboxes with no IPv4 of their own |
| SSH | SSH through a WebSocket tunnel, root by default | SSH gateway with short-lived certificates, paste-anywhere SSH tokens and mosh | SSH into every sandbox over a shared IPv4 host:port |
| Regions | Dedicated servers in the United States | Not published | Germany, Finland and France; no region choice, no US or Asia region; snapshot copies can be stored outside the EU |
Every feature Battleships tracks (43)
| Feature | Agent 37 | Instabox | boat |
|---|---|---|---|
| Full VM (own kernel) | No documented way | Yes | Yes |
| Docker inside | No documented way | Yes | Yes |
| Nested virtualization | No documented way | Not published | Yes |
| Root | Yes | Yes | Yes |
| Systemd | Not published | Yes | Yes |
| Desktop GUI | With your own setup | Yes | Yes |
| Browser | Yes | Yes | Yes |
| Computer-use API | No documented way | Yes | Yes |
| SSH | Yes | Yes | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | No documented way | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Yes | Yes | Yes |
| Memory fork | No documented way | No documented way | No documented way |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | Yes | No documented way |
| Wake on request | Yes | Yes | No documented way |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | Not published | Yes | Yes |
| Public IPv4 | No documented way | No documented way | No documented way |
| Egress allowlist | No documented way | Not published | No documented way |
| Secret proxy | Yes | Yes | Yes |
| Secret proxy for any API | No documented way | Yes | Yes |
| Preinstalled agents | Yes | Yes | Yes |
| Agent harness API | Yes | Not published | Yes |
| MCP server | No documented way | Yes | No documented way |
| GPU | No documented way | Not published | No documented way |
| Windows | No documented way | No documented way | No documented way |
| Arm64 | No documented way | Not published | No documented way |
| Self-hosting | No documented way | Not published | No documented way |
| BYOC | No documented way | Yes | No documented way |
| Open source | No documented way | No documented way | No documented way |
| Scoped API keys | Not published | Yes | Yes |
| Spend limits | Yes | Yes | Yes |
| Usage per sandbox (API) | Yes | Yes | Yes |
| Audit logs | Not published | Yes | No documented way |
| SSO | No documented way | Yes | No documented way |
| SOC 2 | Yes | Not published | No documented way |
| HIPAA | No documented way | Yes | No documented way |
| EU data residency | No documented way | Not published | No documented way |
| Zero data retention | Not published | Not published | Yes |
Performance
Agent 37 and Instabox are 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/s on the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, second of 15. 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. Agent 37 says its default instances share host resources and sells dedicated cores on its Performance type. Instabox says a new machine's shell is ready in under a second; we have not measured it.
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. Agent 37 and Instabox are 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 | Agent 37, a month | Instabox, 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 | $6 · Wallet (any top-up) | $0 · Free | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $12 · Wallet (any top-up) | $25 · Pro | $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 | $329 · Wallet ($50 lifetime top-ups) | $269 · 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 | $33 · Wallet ($1,000 lifetime top-ups) | $500 · Team | $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 | $73 · Wallet ($1,000 lifetime top-ups) · with limits | $500 · Team · with limits | $8,800 · $2000/mo · with limits |
Agent 37: A prepaid wallet with no plan fee, metered per minute: $0.80 per vCPU and $0.70 per GB of RAM per month while running, and $0.09 per GB of disk per month even while stopped or asleep. Performance compute costs 4x and VM compute 2.5x the default rate. New accounts get a one-time $5 credit after adding a card.
Instabox: Free costs $0 for one 4 vCPU / 8 GiB machine. Pro ($20), Builder ($100) and Team ($500) a month each include usage credit equal to the fee. Usage costs $0.004 per vCPU-hour and $0.00225 per GiB-hour, per second while a machine runs. Disk above the included pool costs $0.08 per GiB-month, and egress above the allowance $0.02 per GiB.
Instabox: Plans cost $20, $100, $500 or $2,000 a month, and the whole plan price comes back as VM time for that month. A 4 vCPU / 8 GB sandbox costs $0.036 an hour, per second while it runs. The $20 plan includes 200 starts a day; extra starts cost $0.03 to $0.07. A stopped sandbox has no compute charge and keeps its disk for at least 14 days; extra saved snapshots and the AI credentials proxy still charge. A card is required, also for the 7-day trial.
Limits in those scenarios
In this page's estimates, Agent 37 costs $6, $12, $329 and $33 a month for a workday agent, an always-on agent, 50 agents and 300 user sandboxes. Instabox costs $0, $25, $269 and $500, and boat $20, $26, $317 and $200. The 50-agent fleet needs Docker, so Agent 37 is priced on its VM type there. Agent 37 runs 1,000 awake instances after $1,000 of top-ups, and caps a 4 vCPU / 8 GB default instance at 20 GB of disk. Instabox Team runs 500 machines at once, at 100 starts a minute. boat limits starts: 12 a minute, 60 an hour and 200 a day on the $20 plan, and above 3 times the included rate per minute a start fails with 429. None of the three fits 1,000 short jobs at once as listed: Agent 37 runs that burst on shared vCPUs, Instabox needs more than 500 machines, and boat refuses starts above 270 a minute.
| Scenario | Agent 37 | Instabox | boat |
|---|---|---|---|
| One coding agent, workdays |
| None for this scenario | None for this scenario |
| One agent that never stops |
| None for this scenario | None for this scenario |
| 50 agents in parallel |
| None for this scenario | None for this scenario |
| Sandboxes for your users |
| None for this scenario | None for this scenario |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use Agent 37 when
you host one mostly idle assistant per customer in the US that should sleep with its memory kept, wake on the next message and bill per minute.
- Sleeps and wakes on the next request. For assistants that sit idle most of the day, an instance sleeps with its memory kept and wakes when the next message arrives.
- Low bills for idle and many small sandboxes. Agent 37 bills per minute from a wallet, with no plan fee. Its bill is the lowest of the three for an always-on agent and for 300 user sandboxes.
- Bring a Dockerfile, pick dedicated cores. Agent 37 builds templates from your own Docker image, and sells a Performance type on dedicated cores for steady speed.
Use Instabox when
you want a free machine to start with, Docker on every microVM, pause with memory, machines above 16 vCPU, or mosh over SSH.
- A microVM with Docker on every machine. Every Instabox machine is a Firecracker microVM with its own kernel, root and Docker, with no special type to opt into.
- A free plan and big machines. One 4 vCPU / 8 GiB machine is free, and paid plans go up to 32 vCPU / 64 GiB per machine.
- SSH gateway with mosh. Its SSH gateway relays mosh and gives paste-anywhere SSH tokens, so a terminal session survives sleep and network changes.
Where boat fits
If your agents do hours of real work, like installs, builds, Docker and emulators, look at boat. Every sandbox is a full VM with systemd and /dev/kvm, its disk is saved every minute with no code, and you drive seven installed agents through one API. boat has no free plan, no pause with memory and no US region.
Disk saved every minute, with no code
boat saves the disk every minute while a sandbox runs and once more at each stop. Agent 37 backs up nightly, and on Instabox you save snapshots yourself. If a boat machine fails, you lose what was written after the last good snapshot.
A full Linux server by default
Agent 37's default type cannot run a Docker daemon, and its VM type costs 2.5x with no init system. Instabox does not document /dev/kvm. boat has systemd, Docker and /dev/kvm in every size.
Agents you drive through an API
boat installs seven coding agents that you prompt, steer and follow through one API. On Instabox you run the agent yourself. Agent 37 has a chat API per instance, and every type stops at 4 vCPU.
Use boat when your agents need a full VM with systemd and /dev/kvm, a disk saved every minute with no code, and installed agents you drive through one API. Compare it one to one: boat vs Agent 37 and boat vs Instabox.
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
Questions people ask
Is Agent 37 cheaper than Instabox?
It depends on the workload. An always-on agent costs $12 a month on Agent 37 and $25 on Instabox, and 300 user sandboxes $33 against $500. A workday agent is $0 on Instabox's Free plan against $6. For 50 agents with Docker, Instabox costs $269 and Agent 37 $329. boat costs $20, $26, $200 and $317 for the same four.
Which is faster, Agent 37 or Instabox?
Neither is in the public benchmarks we use, so we cannot say. Instabox says a new shell is ready in under a second, and a paused machine usually resumes in about a second. Agent 37 wakes from a memory checkpoint. We have not measured either.
Can Agent 37 and Instabox run Docker?
Instabox runs Docker and Docker Compose on every machine. Agent 37 runs Docker only on its VM type, which bills compute at 2.5x; its default gVisor type cannot run a Docker daemon. boat runs Docker in every sandbox.
Can I fork a running machine on Agent 37 or Instabox?
Yes, on both. Agent 37 copies the disk of a running, stopped or sleeping instance. Instabox forks a running or stopped machine's disk, up to 5 copies per call. boat forks a running sandbox from its last 1-minute snapshot. None copies memory.
How many machines can I run?
Agent 37 allows 3 awake instances before a top-up and up to 1,000 after $1,000 of top-ups; sleeping and stopped ones do not count. Instabox allows 1 machine on Free, and 100, 200 or 500 running on Pro, Builder or Team. boat allows 100 to 2,000 at once by plan.
Where do Agent 37 and Instabox run?
Agent 37 runs on dedicated servers in the United States. Instabox does not publish its regions. boat runs on machines in Germany, Finland and France, and you cannot pick a region. Snapshot copies can be stored outside the EU.
Try boat
Install the CLI, then run boat new.
7-day free trial