bochka vs NAVER Cloud
NAVER Cloud Server is a general cloud VM in Asia that you set up yourself. bochka is a VM made for AI agents, with coding agents, a 60fps desktop, per-port HTTPS URLs and snapshots every minute ready, and it costs less in the scenarios below.
Where bochka stands out
Coding agents you drive from outside
bochka ships Claude Code, Codex, OpenCode and other agents inside each sandbox, and you drive them from outside with a prompt, steer and events API. A NAVER Cloud server boots a plain OS image, so you install each agent, run it as a process and build the control layer yourself.
A desktop and browser for computer use
Every bochka sandbox can stream a 60fps desktop with Chromium, and agents can click, type and take screenshots through a computer-use API. You can embed the stream in your own app. NAVER Cloud gives you SSH, or RDP on Windows, and no managed desktop.
A sandbox lifecycle built in
bochka snapshots the disk every minute and on stop, a stopped sandbox costs nothing, and you can fork a running sandbox with its disk in one call. Each port can get an HTTPS URL with a token. On NAVER Cloud you take snapshots, make images, attach public IPs and set up TLS yourself.
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
NAVER Cloud
NAVER Cloud is the public cloud of the NAVER group in South Korea; its Server product rents general cloud VMs, not agent sandboxes.
You pick a server type (Standard, High CPU, High Memory, GPU and more), boot it from a Linux or Windows server image, and connect over SSH or RDP. Networking, ACG firewalls, block storage, public IPs and images are separate building blocks you put together in the console or through the API. It is built for companies running general workloads in Korea and the rest of Asia.
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 | NAVER Cloud: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Both run full VMs, but on NAVER Cloud you install the agents, desktop and preview URLs yourself; bochka ships them in every sandbox. | 3rd cheapest of 367 $634 a month | 210th of 367, with limits $2,488 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB A NAVER Cloud server can run all month, but it has no built-in agent API, credentials proxy or per-minute snapshots, so you build that layer yourself. | 5th cheapest of 366 $657 a month | 277th of 366, with limits $5,161 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Both are a compromise for one-minute runs: NAVER Cloud bills by the hour, and bochka's start limits queue big bursts. | 186th of 345, with limits $2,000 a month | 202nd of 345, with limits $10,322 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month 300 servers at once is above what NAVER Cloud confirms by default, and raising a quota needs a business account and a review that can take several days. | Cheapest of 361 $209 a month | 174th of 361, with limits $4,242 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 | 212th of 360, with limits $103 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 | 192nd of 364, with limits $2,121 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 | 343rd of 377, with limits $1,697 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Both fit a dev box that runs all month: a NAVER Cloud server has no session cap, and bochka keeps the disk with snapshots every minute. | 18th cheapest of 368 $26 a month | 78th cheapest of 368 $103 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month bochka streams a 60fps desktop with Chromium and computer use; a NAVER Cloud server has no managed desktop, so you set up VNC or RDP yourself. | 6th cheapest of 369 $200 a month | 327th of 369, with limits $2,828 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 | 351st of 367, with limits $4,242 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 | 302nd of 346, with limits $848 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 | 270th of 338, with limits $16,967 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month bochka runs many isolated sandboxes with fork and snapshots built in; on NAVER Cloud you clone servers from images or replicas by hand. | 2nd cheapest of 360 $200 a month | 300th of 360, with limits $707 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 | bochka | NAVER Cloud |
|---|---|---|
| Kind of product | Sandbox VMs made for AI agents, created by API in seconds | General cloud servers you configure yourself |
| Isolation | Every sandbox is a full VM with its own kernel | Every server is a full VM (KVM on g3, Xen on g1 and g2) |
| Docker and nested virtualization | Docker, root, systemd and /dev/kvm on every machine | You get the whole VM, so you can install Docker; nested virtualization is not documented |
| Coding agents | Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API | None; you install and run agents yourself |
| Desktop and computer use | 60fps desktop and Chromium, computer use and browser control, embeddable stream | No managed desktop; you connect over SSH, or RDP on Windows |
| Snapshots and stops | Snapshots every minute and on stop; a stopped sandbox costs nothing | Server images and block storage snapshots on demand; a stopped server still bills its OS disk |
| Fork | Fork a running sandbox in one call, disk included | Replica servers from a running source reuse its image and settings but not its data; or boot from a saved server image |
| Reaching the sandbox | HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4 | Attach a public IP (billed separately) and open ports in an ACG; TLS is yours to set up |
| Outbound rules | Open internet on every plan; no provider-side allowlist | ACG outbound rules by protocol, port and IP range, up to 50 rules per group (no domain rules) |
| Keeping keys out of the sandbox | AI credentials proxy for model providers and any HTTPS API you register; switch a credential to another account when credits run out | Secret Manager stores secrets that your app reads; no proxy that adds them to requests |
| GPUs | None | NVIDIA A100, H100, L4 and L40S GPU servers |
| Windows | No Windows guests; Wine for Windows apps | Windows server images, with a Windows tool kit |
| Regions | EU only: Germany, Finland, France | Korea, Singapore and Japan for g3; older servers also in the U.S. and Germany |
| 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 | bochka | NAVER Cloud |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | Yes |
| Nested virtualization | Yes | Not published |
| 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 | 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 | No |
| Wake on request | No | No |
| HTTPS preview URLs | Yes | Yes |
| Raw TCP inbound | Yes | Not published |
| Public IPv4 | No | Yes |
| Egress allowlist | No | Yes |
| Secret proxy | Yes | No |
| Secret proxy for any API | Yes | No |
| Preinstalled agents | Yes | No |
| Agent harness API | Yes | No |
| MCP server | No | No |
| GPU | No | Yes |
| Windows | No | Yes |
| 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 | Not published |
| 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: NAVER Cloud was not part of StarSling's public sandbox benchmark. In that run, at 4 vCPU / 8 GB, bochka did 26.69 Node.js build-and-test runs per second on dedicated AMD Ryzen 9 9950X (Zen 5) hosts. NAVER Cloud publishes a fixed network, throughput and IOPS level for each g3 size, but no CPU score we can compare.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. NAVER Cloud 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 | NAVER Cloud, 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 | $25 · Pay as you go |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $103 · Pay as you go |
| 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 | $1,244 · 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 | $4,242 · 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 | $28,278 · Pay as you go · 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.
NAVER Cloud: NAVER Cloud bills each server by the hour at a KRW rate before VAT, with no plan fee, or by a monthly plan you pick in advance. We priced its High CPU g3 server with 4 vCPU / 8 GB, the only size we have, so the two scenarios that use 2 vCPU on bochka use a bigger machine there. Public IPs, extra storage and outbound traffic are billed on top, and a stopped server still bills its OS disk.
Limits in those scenarios
NAVER Cloud has no session cap, so one agent on workdays, one agent all month and 50 agents in parallel all fit. It does not publish how many servers an account can run by default, so 300 servers for your users and 1,000 short jobs at once are not confirmed. Raising a quota needs a business account, and requests above the instant limit can take several days. On bochka the 1,000-job burst hits start limits per minute, hour and day, so it gets queued.
| Scenario | bochka | NAVER Cloud |
|---|---|---|
| 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 NAVER Cloud when
you need GPU servers or Windows, or your data must stay in Korea, Japan or Singapore.
- GPU servers. Useful if your agents train or serve models: NAVER Cloud rents NVIDIA A100, H100, L4 and L40S GPU servers.
- Windows servers. Useful if you need Windows: NAVER Cloud boots Windows server images you reach over RDP.
- Data in Korea, Japan and Singapore. Useful if your users or data must stay in Korea, Japan or Singapore, with IP-level outbound firewall rules.
Use bochka when
you run AI agents that need a ready sandbox: coding agents, a desktop, HTTPS URLs, per-minute snapshots and fork, without building that layer on top of plain VMs.
- Coding agents you drive from outside. On a plain VM you write and maintain the code that starts, prompts and watches every agent, and real AI keys sit on its disk. NAVER Cloud's Secret Manager stores keys but still hands them to the app; bochka's proxy keeps them outside.
- A desktop and browser for computer use. Without it, each computer-use agent needs you to install a desktop, a VNC server and a browser on the server, then build your own screen and input layer and keep it working.
- A sandbox lifecycle built in. A NAVER Cloud replica server copies the image and settings but not the data, so cloning real work means making a server image first. A stopped server keeps billing its OS disk.
Questions people ask
Is bochka cheaper than NAVER Cloud?
In the scenarios we priced, yes: one coding agent on workdays comes to $20 a month on bochka and $25 on NAVER Cloud, and one agent running all month comes to $26 versus $103. 50 agents in parallel come to $317 versus $1,244. NAVER Cloud prices are before VAT, and we only had its 4 vCPU / 8 GB size.
Is NAVER Cloud a sandbox for AI agents?
No. NAVER Cloud Server is a general cloud server: you get a VM with an OS image and build everything else yourself. bochka is made for agents, with coding agents, a desktop, HTTPS URLs and snapshots built into every sandbox.
Can I run Docker on NAVER Cloud and bochka?
Yes on both. A NAVER Cloud server is a full VM, so you can install Docker yourself; nested virtualization is not documented there. On bochka, Docker, systemd, root and /dev/kvm are ready in every sandbox.
Where does my data run?
NAVER Cloud's current g3 servers run in Korea, Singapore and Japan, and its older server lines also run in the U.S. and Germany. bochka runs only in the EU (Germany, Finland and France), with EU data residency and a zero data retention option.
Does bochka have GPUs or Windows like NAVER Cloud?
No. bochka has no GPUs and no Windows guests; it runs Linux VMs on AMD Ryzen 9 9950X hosts and has Wine docs for Windows apps. NAVER Cloud has A100, H100, L4 and L40S GPU servers and Windows server images.
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. NAVER Cloud lists SOC 1, 2 and 3, several ISO/IEC certificates, CSA STAR, PCI DSS, ISMS-P and CSAP.
Start a sandbox
Install the CLI, then run bochka new.
Product preview