boat vs InsForge (InstaCloud)
InstaCloud is a cloud your coding agent deploys apps, Postgres and buckets to from its CLI. boat is the machine the agent works on: a full VM with Docker, a desktop and coding agents inside, whose whole disk survives every stop.
Where boat stands out
The whole machine keeps its disk
On boat the full disk persists through every stop, with snapshots every minute and on stop, and you can fork a running sandbox with everything on it. On InstaCloud only the attached volume survives sleep, and its own changelog says anything outside it does not.
Coding agents inside, driven from outside
boat ships Claude Code, Codex, OpenCode and other agents in each sandbox, and you drive agents inside from outside with a prompt, steer and events API. The AI credentials proxy keeps real keys out of the machine and can switch to another account when credits run out. InstaCloud expects the agent to run on your laptop and injects secrets as environment variables.
A full work machine, not one port
boat gives each sandbox /dev/kvm for emulators, a 60fps desktop with Chromium for computer use, HTTPS on up to 50 ports and raw TCP. An InstaCloud service listens on one HTTP port, cannot expose raw TCP and has no desktop.
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
InsForge (InstaCloud)
InstaCloud, from the InsForge team, is a cloud built for coding agents to deploy to: Postgres, S3-compatible buckets and compute, driven from a CLI.
It is an app platform, not a sandbox. Your coding agent runs on your own machine and uses the insta CLI or MCP server to deploy containers, each into its own microVM with one public HTTPS URL, next to a managed database and storage. Branches give every agent task its own copy-on-write database, forked bucket and running copy of the app. It is for teams who want agents to ship and run backends without a dashboard.
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 | InsForge (InstaCloud): price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month InstaCloud is good at hosting what the agent builds, with a database per branch, but the agent's own tools and packages outside the volume reset on every wake. | 3rd cheapest of 367 $634 a month | 56th of 367, with limits $382 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An assistant that runs with nobody calling it needs always-on on InstaCloud, and only its volume keeps files; on boat the whole VM keeps its disk. | 5th cheapest of 366 $657 a month | 131st of 366, with limits $889 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month InstaCloud fits short runs better: it bills only active CPU and memory per second and has no start-rate limit published, while boat's start limits make big bursts a compromise. | 186th of 345, with limits $2,000 a month | 70th of 345, with limits $85 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month InstaCloud does not publish how many services can run at once beyond 25 of each type per environment on Pro, so 300 user sandboxes needs a talk with them. | Cheapest of 361 $209 a month | 45th of 361, with limits $302 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 | 80th of 360, with limits $4,543 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 | 111th of 364, with limits $370 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month A CI job that runs Docker or emulators fits boat's VM with /dev/kvm; InstaCloud documents Docker inside compute but not nested virtualization. | 4th cheapest of 377 $86 a month | 286th of 377, with limits $500 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Both work for a box you SSH into, but on InstaCloud anything installed outside the volume is gone after it sleeps. | 18th cheapest of 368 $26 a month | 135th of 368, with limits $17 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month InstaCloud has no desktop or browser, so computer-use agents need boat's 60fps desktop or another product. | 6th cheapest of 369 $200 a month | 78th of 369, with limits $108 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month | 7th cheapest of 367 $200 a month | 162nd of 367, with limits $91 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 | 210th of 346, with limits $157 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 | 114th of 338, with limits $3,079 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month Red-team targets often speak raw TCP; InstaCloud exposes only one HTTP port per service, while boat opens raw TCP. | 2nd cheapest of 360 $200 a month | 173rd of 360, with limits $154 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
| Feature | boat | InsForge (InstaCloud) |
|---|---|---|
| Kind of product | A cloud VM for each agent to work on | An app platform agents deploy to: compute, Postgres and buckets |
| Isolation | Every sandbox is a full VM with its own kernel | Every compute service runs in its own microVM |
| Docker, root and nested virtualization | Docker, root, systemd and /dev/kvm on every machine | Root and Docker inside compute since September 2026; no /dev/kvm documented |
| Files after a stop | The whole disk persists, with snapshots every minute and on stop; a stopped sandbox costs nothing | Only the attached volume survives sleep (1 GB by default, up to 10 GB on Pro); the rest resets |
| Sleep and wake | Stop and start by API; resume reboots from disk | Idle services suspend with memory kept and wake on the next request in seconds |
| Fork and branches | Fork a running sandbox, disk included | Branches copy the database (copy-on-write) and bucket; compute restarts from the image with an empty volume |
| Reaching the machine | HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4 | One HTTP port per service on a public HTTPS URL; no raw TCP; SSH with certificates |
| Desktop and computer use | 60fps desktop and Chromium, computer use and browser control | None |
| Coding agents | Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API | Your agent runs on your own machine and drives InstaCloud through the CLI or MCP server |
| Keeping keys out of the machine | AI credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out | Secrets are injected as environment variables at deploy time |
| Managed database and storage | None built in; run any database inside the VM | Managed Postgres, Redis, MySQL and MongoDB, plus S3-compatible buckets |
| Largest machine | 8 vCPU / 16 GB on the $20 plan, 16 vCPU / 32 GB from $100 | 4 vCPU / 4 GB per service on Free, 8 vCPU / 8 GB on Pro and Team |
| Regions | EU only: Germany, Finland, France | US East, Frankfurt and Singapore |
| Self-hosting | Hosted only | Open-source runtime you can run on your own server |
| 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 | InsForge (InstaCloud) |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | Yes |
| Nested virtualization | Yes | No |
| Root | Yes | Yes |
| 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 | Yes |
| 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 | Yes |
| HTTPS preview URLs | Yes | Yes |
| Raw TCP inbound | Yes | No |
| Public IPv4 | No | No |
| Egress allowlist | No | No |
| Secret proxy | Yes | No |
| Secret proxy for any API | Yes | No |
| Preinstalled agents | Yes | No |
| Agent harness API | Yes | No |
| MCP server | No | Yes |
| GPU | No | No |
| Windows | No | No |
| Arm64 | No | No |
| Self-hosting | No | Yes |
| BYOC | No | No |
| Open source | No | Yes |
| Scoped API keys | Yes | Yes |
| Spend limits | Yes | Not published |
| Usage per sandbox (API) | Yes | Yes |
| Audit logs | No | Not published |
| SSO | No | Yes |
| SOC 2 | No | Yes |
| HIPAA | No | Yes |
| EU data residency | Yes | Yes |
| Zero data retention | Yes | Not published |
Performance
There is no comparable measurement for InstaCloud: it was not in the public benchmark run where boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB. boat runs on dedicated AMD Ryzen 9 9950X (Zen 5) hosts.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. InsForge (InstaCloud) 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 | InsForge (InstaCloud), 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 | $10 · InstaCloud Pro · with limits |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $61 · InstaCloud Pro · 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 | $762 · InstaCloud Pro · 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 | $223 · InstaCloud Free · 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 | $792 · InstaCloud Free · 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 whole disk.
InsForge (InstaCloud): InstaCloud bills per second for active CPU and memory, plus volume storage and $0.05 per GB of egress. Pro is a $20 monthly minimum that becomes usage credit; Free gives $10 a month and pauses projects when it runs out; Team adds a $499 platform fee on top of usage.
Limits in those scenarios
InstaCloud's prices in these scenarios come with a catch: only the volume keeps files across a sleep, so an agent's installed tools reset unless they live there. It does not publish how many services can run at once beyond 5, 25 or 100 of each type per environment by plan, so 50 agents, 300 user sandboxes and 1,000 jobs are not confirmed. 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 | InsForge (InstaCloud) |
|---|---|---|
| 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 InsForge (InstaCloud) when
your agent's job is to ship and run a backend: an app with Postgres and buckets, a branch per task, scale to zero, in the US or Asia or on your own server.
- Database, storage and branches built in. Useful if your agent ships backends: each task branch gets its own copy-on-write Postgres, forked bucket and live URL.
- Sleeps with memory and wakes on a request. Useful for apps that sit idle: a service suspends with its memory and wakes in seconds when a request arrives.
- US and Asia regions, and self-hosting. Useful if you serve users outside Europe or must run on your own server: three regions and an open-source runtime.
Use boat when
the agent needs a real machine to work on: a full VM whose whole disk persists, Docker and /dev/kvm, a desktop, many ports and coding agents you drive from outside.
- The whole machine keeps its disk. An agent installs packages, toolchains and caches all over the filesystem. On InstaCloud those vanish on each wake unless you move them onto a volume of 1 GB by default, 10 GB at most on Pro.
- Coding agents inside, driven from outside. Running many agents means many laptops or your own orchestration. Keys in environment variables can be read by any code the agent runs, and when an account runs out of credits the agent stops.
- A full work machine, not one port. An agent that tests an Android app, drives a browser or runs a dev server next to a database admin port hits a wall on one HTTP port with no screen.
Questions people ask
Is boat cheaper than InstaCloud?
It depends on the scenario. One coding agent on workdays comes to $20 a month on boat and $10 on InstaCloud, but one agent running all month comes to $26 versus $61, and 50 agents in parallel to $317 versus $762. On InstaCloud those prices keep only the volume across sleeps, and none of its scenarios fully fit.
Is InstaCloud a sandbox?
Not in the usual sense. It is an app platform: your coding agent runs elsewhere and deploys containers, databases and buckets to it, each container in its own microVM. boat is the machine the agent itself runs on.
Do files survive when an InstaCloud service sleeps?
Only files on the attached volume, which is 1 GB by default and up to 10 GB on Pro. InstaCloud's changelog says anything outside the volume does not survive sleep and wake. On boat the whole disk persists, with snapshots every minute.
Can I run Claude Code inside InstaCloud?
InstaCloud is built for Claude Code, Cursor or Codex on your own machine to drive it through a CLI or MCP server. You could install an agent inside a compute service yourself, but there is no API to drive it. boat preinstalls coding agents and gives you a prompt, steer and events API.
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. InstaCloud runs in US East, Frankfurt and Singapore, and you can self-host its open-source runtime.
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. InstaCloud lists SOC 2 and SSO on its Team plan.
Start a sandbox
Install the CLI, then run boat new.
7-day free trial