bochka vs Buddy Sandboxes
Buddy Sandboxes are persistent Ubuntu machines that live next to Buddy's CI/CD pipelines. bochka gives every sandbox its own VM with a 60fps desktop, snapshots every minute and coding agents you drive from outside with an API.
Where bochka stands out
Coding agents you drive from outside
bochka ships Claude Code, Codex, OpenCode and other agents in each sandbox, and you drive them from your own code with a prompt, steer and events API. Buddy preinstalls Claude Code too, but you start it in the terminal, and its SDK and API manage the machine rather than the agent. bochka's credentials proxy also keeps real AI keys out of the sandbox.
A desktop and a browser in every sandbox
Every bochka sandbox has a 60fps desktop and Chromium, with computer use and browser control, and you can embed the stream in your own app. Buddy Sandboxes are reached by terminal, SSH and tunnels, and document no desktop or browser.
Snapshots every minute, without asking
bochka snapshots each sandbox's disk every minute and again on stop, so you can roll back to a recent point after a bad change. Buddy keeps the disk too, but takes snapshots only when you, a pipeline or a script inside the sandbox ask for one, and they pile up until you delete them.
The companies and products
bochka
bochka.xyz 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
Buddy Sandboxes
Buddy is a CI/CD and delivery platform from 2015; Sandboxes are its newer product for long-lived Linux machines.
A Buddy Sandbox is an isolated Ubuntu machine in Buddy's cloud with passwordless sudo, Docker, a 50 GB persistent disk and Claude Code preinstalled. You run it from the web terminal, SSH, a TypeScript SDK, a REST API, a CLI or an MCP server, and publish apps through tunnel endpoints. It is built for teams that already use Buddy pipelines and want dev boxes, preview environments and agent machines in the same workspace.
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 | bochka: price rank | Buddy Sandboxes: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Buddy gives agents a Linux machine with Docker and Claude Code, but no desktop or browser, and nested virtualization is not documented, so app builds that need a screen or an emulator are a compromise there. | 3rd cheapest of 367 $634 a month | 196th of 367, with limits $2,245 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB | 5th cheapest of 366 $657 a month | 227th of 366, with limits $2,792 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Buddy bills short runs by the CPU-minute with no plan fee in this model, so quick bursts come out ahead there, while bochka's start limits make big bursts a compromise. | 186th of 345, with limits $2,000 a month | 96th of 345, with limits $146 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Buddy does not publish how many sandboxes can run at once, so 300 sandboxes for your users is not confirmed there. | Cheapest of 361 $209 a month | 100th of 361, with limits $919 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 | 76th of 360, with limits $4,375 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 | 135th of 364, with limits $700 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Both fit; Buddy also runs the CI pipelines themselves, while bochka gives each job a full VM with Docker and /dev/kvm. | 4th cheapest of 377 $86 a month | 55th cheapest of 377 $425 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Both fit: Buddy has no session limit and keeps the disk, and bochka keeps the disk with snapshots every minute. | 18th cheapest of 368 $26 a month | 84th cheapest of 368 $112 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Buddy documents no desktop, browser or computer-use interface, so computer-use agents need you to build that layer yourself. | 6th cheapest of 369 $200 a month | 195th of 369, with limits $642 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month bochka ships Chromium and a desktop in every sandbox; on Buddy you would install and run a headless browser yourself. | 7th cheapest of 367 $200 a month | 237th of 367, with limits $350 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 | 232nd of 346, with limits $213 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 | 145th of 338, with limits $4,265 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 | 206th of 360, with limits $214 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 | bochka | Buddy Sandboxes |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | Yes |
| Nested virtualization | Yes | Not published |
| 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 | No |
| 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 | Yes |
| 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 | Yes |
| GPU | No | No |
| 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 | Yes |
| SOC 2 | No | Yes |
| HIPAA | No | No |
| EU data residency | Yes | Yes |
| Zero data retention | Yes | Not published |
Performance
There is no comparable measurement for Buddy Sandboxes: they were not in the public StarSling hpc-sandbox-benchmarks run. In that run, bochka did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB 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. Buddy Sandboxes 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 | bochka, a month | Buddy Sandboxes, 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 | $31 · Free |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $117 · Free |
| 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 | $1,449 · Free |
| 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 | $394 · 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 | $875 · Free · with limits |
bochka: 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.
Buddy Sandboxes: Buddy charges per CPU-minute and per GB-hour beyond what your plan includes; every plan includes only 300 sandbox CPU-minutes a month. Plan fees are in euros (Pro EUR 29, Hyper EUR 99) while sandbox rates are in dollars, and the pricing page does not say what a GB-hour measures. The Buddy figures here use the published rates with no plan fee, but Buddy's plan docs say Sandboxes start on Pro.
Limits in those scenarios
Buddy does not publish how many sandboxes can run at once, so 300 sandboxes for your users and 1,000 short jobs at once are not confirmed there, and it does not publish an egress price. bochka's 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. One Buddy sandbox tops out at 12 CPU / 24 GB with a fixed 50 GB disk.
| Scenario | bochka | Buddy Sandboxes |
|---|---|---|
| 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 Buddy Sandboxes when
your team already runs CI/CD on Buddy, or you want preview machines that sleep when idle and wake on traffic, in the US or Asia.
- Sleeps when idle, wakes on traffic. Useful for preview apps that get visits now and then: a Buddy Sandbox stops when idle and starts again when a request arrives.
- CI/CD in the same place. Useful if your team already ships with Buddy pipelines: sandboxes sit in the same workspace and pipelines can create and snapshot them.
- US and Asia regions. Useful if your users or data sit in the US or Asia: Buddy runs sandboxes and endpoints in the US, the EU and Asia.
Use bochka when
agents do the work: coding agents you drive by API, a desktop and browser for computer use, snapshots every minute and keys that never enter the machine.
- Coding agents you drive from outside. On Buddy, running many agents means writing your own code to start each one over a command, feed it prompts and read its output, and the real API key sits inside the machine as a variable.
- A desktop and a browser in every sandbox. An agent that tests a web app, fills a form or clicks through a GUI needs a screen. On Buddy you would install a display server, a browser and a remote viewer yourself and keep them working.
- Snapshots every minute, without asking. Agents break things without warning. On Buddy, if nobody called snapshot create before the agent wiped a folder, there is no earlier state to restore.
Questions people ask
Is bochka cheaper than Buddy Sandboxes?
In most of the scenarios we priced, yes: one coding agent on workdays comes to $20 a month on bochka and $31 on Buddy, one agent running all month to $26 versus $117, and 50 agents in parallel to $317 versus $1,449. For 1,000 short jobs at once, Buddy comes to $875 against bochka's $2,000, and neither is confirmed to fit.
Do Buddy Sandboxes come with Claude Code?
Yes. Buddy preinstalls Claude Code, and you start it from the terminal or over SSH. Its SDK, REST API and MCP server create and manage the machine. bochka preinstalls Claude Code, Codex and other agents and adds an API to send prompts, steer the agent and read its events from outside.
Can I use Buddy Sandboxes without a paid plan?
The pricing page lists 300 sandbox CPU-minutes and 730 GB-hours on the Free plan, but Buddy's plan docs say Free covers everything except 2FA and Sandboxes, and that Pro is the first plan with them. Check with Buddy before you plan around the Free plan.
Do Buddy Sandboxes keep files and memory after a stop?
The disk is kept, but memory is not: running processes end when the sandbox stops, and configured apps start again on the next boot. bochka works the same way for memory, and it also snapshots the disk every minute and on stop.
Is bochka SOC 2 compliant?
SOC 2 is in progress. We're open about where we stand: live status at status.bochka.xyz and our data retention policy at docs.bochka.xyz/data-retention. Buddy's security page lists SOC 2 Type II, PCI DSS, GDPR and CCPA.
Start a sandbox
Install the CLI, then run bochka new.
Product preview