boat vs Tencent Cloud Studio
Tencent Cloud Studio is a cloud IDE you open in a browser tab, with GPUs and the CodeBuddy assistant built in. boat gives each agent its own full VM that keeps running without a tab, with SSH, hosting and coding agents you drive from outside.
Where boat stands out
A full VM, not an IDE workspace
On boat every sandbox has its own kernel, with root, systemd, Docker and /dev/kvm ready. Cloud Studio gives you a workspace with Docker built in, but its docs say nothing about a VM per workspace, root or systemd, and there is no /dev/kvm.
Runs for days with no tab open
A boat sandbox keeps running until you stop it, with no session cap on paid plans, and you reach it by SSH, HTTPS on up to 50 ports or raw TCP. A Cloud Studio workspace shuts down at its timer, 24 hours at most, and a person has to re-enter it to renew. Its ports are preview links only, with no SSH, TCP or UDP.
Coding agents you drive from outside
boat ships Claude Code, Codex, OpenCode and other agents inside each sandbox, and you drive agents inside from outside with a prompt, steer and events API. Cloud Studio's CodeBuddy works inside the editor, for a person at the keyboard. boat's credentials proxy also keeps real AI keys out of the sandbox, while Cloud Studio injects secrets as environment variables.
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
Tencent Cloud Studio
Tencent Cloud Studio is Tencent Cloud's browser-based IDE: you write and run code in a workspace from a web page, with nothing to install.
Each workspace has its own storage, compute and a ready-made stack such as Python, Node.js or Java, and the new version has Docker built in. You can pick CPU sizes or GPUs, share an app with collaborators, fork it, and use the CodeBuddy coding assistant inside the editor. It is built for developers, students and teachers on the Chinese market, so it is a different kind of product from an agent sandbox; the two overlap where you want a cloud Linux machine for coding work.
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 | Tencent Cloud Studio: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Cloud Studio is only a compromise here: a workspace stops after at most 24 hours unless a person renews the timer, and there is no SSH or raw TCP. | 3rd cheapest of 367 $634 a month | 101st of 367, with limits $898 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An assistant that runs all month needs a machine that never stops; Cloud Studio shuts down at its timer, while boat has no session cap on paid plans. | 5th cheapest of 366 $657 a month | 158th of 366, with limits $1,169 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Cloud Studio is cheaper for short bursts of code runs, but it has no API to run code in a workspace, so you would build that part yourself. | 186th of 345, with limits $2,000 a month | 210th of 345, with limits $49 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Cloud Studio's API can create and stop workspaces for your users, but it does not publish how many can run at once. | Cheapest of 361 $209 a month | 206th of 361, with limits $383 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 | 242nd of 360, with limits $1,852 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 | 241st of 364, with limits $298 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month | 4th cheapest of 377 $86 a month | 250th of 377, with limits $179 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Both give you a cloud dev machine; boat adds SSH, root and a box that keeps running with no browser tab open. | 18th cheapest of 368 $26 a month | 223rd of 368, with limits $47 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Cloud Studio has no desktop or browser for an agent to drive; boat streams a 60fps desktop with Chromium and computer use. | 6th cheapest of 369 $200 a month | 119th of 369, with limits $268 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 | 175th of 367, with limits $134 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 | 163rd of 346, with limits $73 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month Cloud Studio ranks a little higher here, mostly because its small sizes cost less per hour, but it has no SSH, raw TCP or public IP. | 71st of 338, with limits $1,800 a month | 55th of 338, with limits $1,453 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 | 133rd of 360, with limits $88 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month Only Cloud Studio has GPUs; boat has none. | 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 | Tencent Cloud Studio |
|---|---|---|
| What it is | An API and CLI for full Linux VMs that agents and programs run in | A browser IDE; you work in a workspace through a web page |
| Isolation | Every sandbox is a full VM with its own kernel | Workspaces; no VM or own kernel per workspace is documented |
| Docker, root and nested virtualization | Docker, root, systemd and /dev/kvm on every machine | Docker built in on the new version; root and systemd not documented; no /dev/kvm |
| How long it runs | No session cap on paid plans; runs with nothing open | Auto-shutdown at 1 hour by default, 24 hours at most; a person must re-enter to renew it |
| Files after a stop | Snapshots every minute and on stop; a stopped sandbox costs nothing | Storage is kept after a stop until you delete the workspace; no snapshots |
| Fork | Fork a running sandbox | Fork an app from its card to make a copy |
| Reaching the machine | SSH; HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4 | Preview links per port, public or only you; WebSocket works; no SSH, TCP or UDP |
| Desktop and computer use | 60fps desktop and Chromium, computer use and browser control | None; graphical programs are not supported yet |
| Coding agents | Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API | CodeBuddy inside the editor, 500 free credits a month; no API to drive it |
| Keeping keys out of the machine | AI credentials proxy for model providers and any HTTPS API you register | Secrets are environment variables injected into the workspace |
| API | Create, run commands, fork, snapshot, host and drive agents from code | Create, change, run, stop, delete and list workspaces; free high-performance workspaces have no API |
| GPUs | None | T4, V100, A10, L40 and A800 by the hour |
| Sizes | Up to 16 vCPU / 32 GB, by plan | From 1 vCPU / 2 GB up to 64 vCPU / 128 GB, switchable inside the workspace |
| Regions and accounts | EU only: Germany, Finland, France | Tencent Cloud in China; some compute needs a Chinese ID card check |
| 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 | Tencent Cloud Studio |
|---|---|---|
| Full VM (own kernel) | Yes | No |
| 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 | No |
| Persistent disk | Yes | Yes |
| Pause/resume | Yes | Yes |
| Automatic snapshots | Yes | No |
| 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 | 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 | Yes |
| Agent harness API | Yes | No |
| MCP server | No | No |
| GPU | No | Yes |
| Windows | No | No |
| Arm64 | No | No |
| Self-hosting | No | No |
| BYOC | No | No |
| Open source | No | No |
| Scoped API keys | Yes | Not published |
| Spend limits | Yes | Not published |
| Usage per sandbox (API) | Yes | No |
| Audit logs | No | Not published |
| SSO | No | No |
| 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 Cloud Studio: it was not part of StarSling's public sandbox benchmark. boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB in that benchmark, 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. Tencent Cloud Studio 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 | Tencent Cloud Studio, 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 | $12 · Pay as you go |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $49 · Pay as you go · 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 | $590 · Pay as you go |
| 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 | $168 · Pay as you go · 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 | $370 · Pay as you go · 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.
Tencent Cloud Studio: Cloud Studio bills machine hours per size, in yuan, from a prepaid pack or pay as you go from your Tencent Cloud balance. Billing runs while an IDE tab is open, even with nothing running, and for about 10 more minutes after you close it; when a pack runs out, the app keeps running on pay as you go. Disk and network prices are not published.
Limits in those scenarios
The limit that bites on Cloud Studio is time: a workspace shuts down at its timer, 24 hours at most, so one agent that never stops does not fit. It also does not publish how many workspaces can run at once, so 300 sandboxes for your users and 1,000 short jobs at once 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 | Tencent Cloud Studio |
|---|---|---|
| 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 Tencent Cloud Studio when
people in China want a browser IDE with GPUs and CodeBuddy, for coding by hand, teaching or model work.
- GPUs in the same workspace. Useful if you train or test models: Cloud Studio rents T4 to A800 GPUs by the hour inside the same IDE.
- A cloud IDE for people in China. Useful if your team codes from mainland China: Cloud Studio runs on Tencent Cloud with CodeBuddy and live collaboration in the browser.
- Small sizes up to very big ones. Useful for light coding or very large builds: sizes run from 1 vCPU / 2 GB up to 64 vCPU / 128 GB.
Use boat when
agents work on their own for hours or days: full VMs with root, Docker and /dev/kvm, SSH and hosting, a desktop, and coding agents you drive from your code.
- A full VM, not an IDE workspace. Agents install system packages, run services under systemd and start emulators that need /dev/kvm. On Cloud Studio you find out by trial what is allowed, and emulators are out.
- Runs for days with no tab open. An agent job that runs overnight past the timer is killed at shutdown with no second warning. Without SSH or TCP you cannot attach tools, databases or game servers from outside.
- Coding agents you drive from outside. With CodeBuddy someone has to sit in the browser and type. To run 50 agents from your own code you need an API to start, prompt and watch them, and keys an agent cannot read.
Questions people ask
Is boat cheaper than Tencent Cloud Studio?
It depends on the job. One coding agent on workdays comes to $20 a month on boat and $12 on Cloud Studio, and Cloud Studio is also lower for sandboxes for your users and short bursts, though its concurrency there is not confirmed. One agent running all month comes to $26 on boat and $49 on Cloud Studio, which does not fit because of its 24 hour cap, and 50 agents in parallel come to $317 versus $590.
Can I SSH into Tencent Cloud Studio?
No. Its FAQ says direct SSH is not supported, and port forwarding covers HTTP and WebSocket, not TCP or UDP. boat gives every sandbox SSH, HTTPS on up to 50 ports and raw TCP.
How long can a Cloud Studio workspace run?
Auto-shutdown starts at 1 hour and can be set to 24 hours at most. To run longer, a person must re-enter the workspace and renew the timer, or the work stops at shutdown. boat has no session cap on paid plans.
Can I use Tencent Cloud Studio from outside China?
You need a bound Tencent Cloud account, and the real-name check that unlocks some compute asks for a Chinese ID card number. boat runs in the EU (Germany, Finland and France), with EU data residency and a zero data retention option.
Does boat have GPUs like Cloud Studio?
No. boat runs Linux VMs on AMD Ryzen 9 9950X machines and has no GPUs. Cloud Studio rents T4, V100, A10, L40 and A800 GPUs by the hour.
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