bochka vs Together Code Sandbox

Together Code Sandbox starts VMs from Docker-image snapshots and can save their memory when they end. bochka gives each agent a whole machine you can sign up for today, with SSH, Docker, /dev/kvm, private HTTPS hosting, a desktop and coding agents built in.

Where bochka stands out

A whole machine, not just a command runner

A bochka sandbox is a full VM with SSH, systemd, root, Docker and /dev/kvm, so Android emulators and nested VMs run too. Together's new SDK runs commands as a normal user by default, does not read Dev Containers or Docker Compose, and documents neither Docker inside nor nested virtualization.

Hosting that is private by default

bochka serves HTTPS on up to 50 ports per sandbox, token-protected by default, and also takes raw TCP on a dedicated IPv6 or IPv4. On Together every port is public, preview URLs are listed as coming soon, and access rules are an experimental feature.

Agents, desktop and keys built in

Claude Code, Codex and other agents come preinstalled, and you prompt, steer and follow them from outside through one API. Each sandbox also has a 60fps desktop with Chromium for computer use, and real API keys stay outside it in a credentials proxy that can switch to another account when one runs out of credits.

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
Hardware: Dedicated AMD Ryzen 9 9950X (Zen 5) hosts
Regions: EU: Germany, Finland, France
Plans: From $20 a month, all of it usage credit, 7-day trial

Together Code Sandbox

Together AI is a cloud for running and training AI models, and Code Sandbox is its product for running code in isolated VMs.

You build a snapshot from a Docker image or Dockerfile on Together's builder, then start sandboxes from it and run commands and file operations through the together-sandbox SDK or CLI. When a sandbox ends you can save its disk, and its memory too, as a new snapshot to start more sandboxes from. It is aimed at model-generated code, agent tool calls and reinforcement learning rollouts, and it must be enabled for your organization by Together. Together also sells Code Interpreter, a separate API that runs Python code in short sessions.

Company: Together AI, a cloud for model inference, fine-tuning and GPU clusters
Isolation: A virtual machine per sandbox on Together's infrastructure
Access: The SDK and CLI must be enabled for your organization; you ask Together for access
Machine size: 0.1 to 16 cores, 1 to 8 GB of memory per core; disk size is not configurable
History: Built on CodeSandbox, now a Together company; the CodeSandbox-based SDK is deprecated and together-sandbox is its successor
Source code: The SDKs and CLI are MIT-licensed on GitHub; the platform itself is not

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	Together Code Sandbox: price rank
Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A software factory needs Docker, root and often /dev/kvm; bochka has all three on every machine, while Together's new SDK runs commands as a normal user and does not document Docker inside or nested virtualization.	3rd cheapest of 367 $634 a month	65th of 367, with limits $528 a month
Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An assistant that lives on its machine all month fits bochka; Together sandboxes have no idle detection or wake on request, so you start and stop them from your own backend.	5th cheapest of 366 $657 a month	152nd of 366, with limits $1,095 a month
Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Together's Code Interpreter suits short Python runs, billed per 60-minute session; it runs only Python and keeps nothing after the session ends.	186th of 345, with limits $2,000 a month	326th of 345, with limits $2,190 a month
Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month On Together every port of a sandbox is public and preview URLs are listed as coming soon, so giving each user a private app needs work that bochka's token-protected HTTPS URLs already do.	Cheapest of 361 $209 a month	234th of 361, with limits $900 a month
RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month Together can save a sandbox with its memory and start many copies from it, which suits rollouts that branch from one warm state, but the VM product needs Together to enable it.	34th of 360, with limits $2,000 a month	315th of 360, with limits $21,900 a month
Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month Together documents a Harbor integration for running agent evals; bochka ships the coding agents themselves, with an API to prompt, steer and follow each run.	10th cheapest of 364 $200 a month	258th of 364, with limits $450 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	251st of 377, with limits $180 a month
24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB	18th cheapest of 368 $26 a month	210th of 368, with limits $22 a month
Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Computer-use agents need a desktop and a browser, which bochka streams at 60fps and Together Code Sandbox does not offer.	6th cheapest of 369 $200 a month	184th of 369, with limits $600 a month
Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month bochka ships Chromium on a full desktop in every sandbox; on Together you install a headless browser in your own image.	7th cheapest of 367 $200 a month	301st of 367, with limits $900 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	219th of 346, with limits $180 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	124th of 338, with limits $3,600 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	167th of 360, with limits $150 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	Together Code Sandbox
Isolation	Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts	A VM per sandbox on Together's infrastructure
Getting started	Sign up and start a sandbox from the API, CLI or dashboard	The SDK and CLI must be enabled for your organization by Together
Machine size	2 to 16 vCPU with 2 GB per vCPU, up to 16 vCPU / 32 GB on the $100 plan and above	0.1 to 16 cores, 1 to 8 GB of memory per core; disk size fixed
Docker, root and systemd	Docker, systemd and root on every machine	Commands run as user 1000 by default, root on request; Dev Containers and Docker Compose are not read; Docker inside not documented
Nested virtualization	Yes, /dev/kvm for Android emulators and VMs	Not documented
Shell access	SSH, plus exec and files over the API	Commands, interactive shells and files through the SDK and CLI; no SSH
Saving state	Disk saved every minute and on stop; a stopped sandbox costs nothing; resume boots the machine again	End a sandbox with a snapshot of its disk, or disk and memory; a new sandbox resumes with its processes running
Fork	Fork a running sandbox; the original keeps running	No live fork: the parent stops, then you start many sandboxes from its snapshot
Idle and wake	Runs until you stop it or its time to live ends; started again by API	Time to live counts from creation; no idle detection and no wake on request
Reaching the sandbox	HTTPS URLs for up to 50 ports, token-protected by default; raw TCP; dedicated IPv6 or IPv4	All ports public; preview URLs coming soon; an experimental inbound allowlist by IP or token
Desktop and browser	Linux desktop streamed at 60fps, Chromium and computer-use tools	None; headless command and file access only
Coding agents	Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API	Run your own agent through execs; a Harbor guide for agent evals
Keeping API keys out of the sandbox	Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out	No credentials proxy documented
Hosted code interpreter	No separate product; run code through exec in a sandbox	Code Interpreter API runs Python in sessions billed per 60 minutes
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	Together Code Sandbox
Full VM (own kernel)	Yes	Yes
Docker inside	Yes	Not published
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	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	Yes
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	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	No
BYOC	No	No
Open source	No	No
Scoped API keys	Yes	Not published
Spend limits	Yes	Yes
Usage per sandbox (API)	Yes	Yes
Audit logs	No	Not published
SSO	No	Yes
SOC 2	No	Yes
HIPAA	No	No
EU data residency	Yes	Not published
Zero data retention	Yes	Not published

Performance

bochka26.69 runs/s

Together Code SandboxNot measured

There is no comparable CPU measurement for Together Code Sandbox. bochka did 26.69 runs per second on the StarSling hpc-sandbox-benchmarks Node.js test at 4 vCPU / 8 GB, on dedicated AMD Ryzen 9 9950X hosts, and Together was not part of that run.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Together Code Sandbox 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	Together Code Sandbox, 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	$5 · Together pay-as-you-go (prepaid credits; Code Interpreter only) · with limits
One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month	$26 · $20/mo	$22 · Together pay-as-you-go (prepaid credits; Code Interpreter only) · 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	$264 · Together pay-as-you-go (prepaid credits; Code Interpreter only) · 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	$900 · Together pay-as-you-go (prepaid credits; Code Interpreter only) · 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	$6,000 · Together pay-as-you-go (prepaid credits; Code Interpreter only) · with limits

bochka: Plans start at $20 a month, and the whole fee becomes usage credit. You pay per second while a sandbox runs, and a stopped sandbox costs nothing and keeps its disk.

Together Code Sandbox: Together lists Code Sandbox at $0.0446 per vCPU-hour and $0.0149 per GiB-hour, but the VMs must be enabled for your organization, so the table prices Together with Code Interpreter, its only self-serve product with a published price: $0.03 per 60-minute session from prepaid credits. Code Interpreter runs only Python and keeps no disk, so it does not fit these scenarios.

Limits in those scenarios

The Together column is Code Interpreter, and its limits decide every row: sessions last 60 minutes, nothing is kept after a session, it runs only Python, and the machine size is not published. Together's VM product is closer to bochka, with sandboxes up to 16 cores and a fixed disk, but it needs Together to turn it on. On bochka the limit that bites is starts: 1,000 short jobs at once need more starts per hour and per day than even the $2,000 plan allows, so starts are queued.

Scenario	bochka	Together Code Sandbox
One coding agent, workdays	None that bite	Together pay-as-you-go (prepaid credits; Code Interpreter only): sessions capped at 1 h missing persistent disk machine size not published different kind of product: code interpreter
One agent that never stops	None that bite	Together pay-as-you-go (prepaid credits; Code Interpreter only): sessions capped at 1 h missing persistent disk machine size not published different kind of product: code interpreter
50 agents in parallel	None that bite	Together pay-as-you-go (prepaid credits; Code Interpreter only): sessions capped at 1 h missing persistent disk, Docker inside machine size not published different kind of product: code interpreter
Sandboxes for your users	None that bite	machine size not published different kind of product: code interpreter machine size per session not published egress price not published
1,000 short jobs at once	max 2,400 starts/day (needs ~6,667) max 600 starts/hour (needs ~1,000) bursts above 90 starts/min are queued (a burst of 333 takes ~4 min to start)	machine size not published different kind of product: code interpreter machine size per session not published egress price not published

Which one should you use?

Use Together Code Sandbox when

you already run models on Together and want light sandboxes, from a tenth of a core, that you can save with their memory and branch from, or a simple Python runner billed per session.

Memory snapshots to branch from. Useful for rollouts and test variants: save a sandbox with its memory and start many copies with processes still running.
Sandboxes next to your models. Handy if you already run inference or fine-tuning on Together: sandboxes and models share one account and API key.
Very small machines and a Python runner. Good for many light jobs: sandboxes start at a tenth of a core, and a Python interpreter API bills per session.

Use bochka when

your agents do real work for hours: installs, builds, Docker, emulators, a browser and servers other people reach, on a full machine you can start today with the agent and its keys already set up.

A whole machine, not just a command runner. Agents and their tools expect a shell, a service manager and often KVM. Without them, compose stacks, emulator builds and VM-based tests do not start, and each missing tool has to be baked into the image first.
Hosting that is private by default. An agent that starts a dev server or a database on Together exposes it to anyone who finds it until you write an experimental policy, and there is no stable preview URL yet to hand to a user.
Agents, desktop and keys built in. On Together you build the harness, browser and desktop into your own image and wire every step through execs. Keys you pass in sit inside a machine the agent controls, where one prompt injection can leak them.

Questions people ask

Is bochka cheaper than Together Code Sandbox?

The only self-serve Together price is Code Interpreter, and it does not fit these scenarios because sessions end after 60 minutes and keep no disk. It looks cheaper for small runs, $264 against $317 a month for 50 agents, but costs more at scale: $900 against $200 for sandboxes for your users, and $6,000 against $2,000 for 1,000 short jobs at once.

What is the difference between Together Code Sandbox and Code Interpreter?

Code Sandbox gives you VMs started from Docker-image snapshots, with commands, files and memory snapshots, and must be enabled for your organization. Code Interpreter is a simpler API that runs Python code in a session, billed per 60 minutes.

Is Together Code Sandbox the same as CodeSandbox SDK?

Not anymore. The first version ran on CodeSandbox's microVMs and is now called deprecated in Together's docs. Its successor, the together-sandbox SDK, drops live fork, idle hibernation and wake on request, and replaces them with snapshots taken when a sandbox ends.

Can I run Docker and Android emulators in each?

bochka runs Docker, systemd and root on every machine and exposes /dev/kvm for Android emulators and nested VMs. Together builds sandboxes from Docker images, but its new SDK documents neither Docker inside a sandbox nor nested virtualization.

Can bochka save a sandbox with its memory like Together?

No. bochka saves the disk every minute and when the sandbox stops, so files survive and a stopped sandbox costs nothing, but resuming boots the machine again. Together can save memory too, so a new sandbox comes back with its processes running.

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.

Start a sandbox

Install the CLI, then run bochka new.

Product preview