bochka vs Claude Managed Agents

Claude Managed Agents is Anthropic's hosted agent: you send a task, and Claude works in a small container that Anthropic runs. bochka gives each agent a whole Linux VM that you can SSH into, open in a browser and keep, with Claude Code, Codex and other agents that you drive from outside.

Where bochka stands out

A whole machine, not a small container

A bochka sandbox is a full VM with root, Docker, systemd and /dev/kvm, up to 16 vCPU / 32 GB, and its disk is saved every minute for as long as you keep it. A Managed Agents sandbox is a container with up to 8 GB of memory and 10 GB of disk, Docker in limited availability, and state that expires 30 days after the sandbox is created.

Open what the agent builds

bochka gives each sandbox HTTPS URLs on up to 50 ports, token-protected by default, plus raw TCP, SSH and a Linux desktop streamed at 60fps. A Managed Agents sandbox has no inbound ports or SSH; you see its work only through the events API and the files it writes out.

Any agent, driven from outside

bochka has Claude Code, Codex and five other agents preinstalled, and you prompt, steer and follow them through one API. Real API keys stay outside the sandbox in a credentials proxy, and a credential 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

Claude Managed Agents

Claude Managed Agents is Anthropic's ready-made agent harness: Claude runs as an autonomous agent in a cloud sandbox that Anthropic manages, and you talk to it through an events API.

This is an agent platform, not a sandbox you rent. You define an agent (model, prompt, tools, MCP servers, skills) and an environment, then start sessions, send messages as events and stream back what Claude does; the harness handles tool calls, prompt caching and compaction. Each session gets its own Ubuntu container, or runs its tools on a worker you host yourself. It is built for teams that want Claude doing long tasks without building their own agent loop or sandbox.

Isolation: Each session gets its own isolated Linux container on Anthropic-managed infrastructure
Sandbox size: Ubuntu 24.04, x86_64, up to 8 GB memory and 10 GB disk; vCPU count not published
Models: Claude models only, through Anthropic's own harness
Billing: Model tokens at API rates, plus a session runtime fee that accrues only while a session is running
Self-hosted sandboxes: Tools can run on a worker in your own infrastructure while Anthropic keeps the orchestration
Data location: Workspace geo, which sets where data is stored and code runs, is US only today; inference can be global or US
Status: Beta, behind the managed-agents-2026-04-01 header; not eligible for Zero Data Retention or a HIPAA BAA

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	Claude Managed Agents: price rank
Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A coding agent on Managed Agents works in a container with up to 8 GB of memory and Docker only in limited availability; on bochka it gets a full VM with Docker, systemd and a URL to open the app it builds.	3rd cheapest of 367 $634 a month	106th of 367, with limits $964 a month
Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Managed Agents is built for long sessions that sleep between messages for free, but sandbox files are kept only 30 days after the sandbox is created; a bochka disk stays as long as you keep the sandbox.	5th cheapest of 366 $657 a month	219th of 366, with limits $2,511 a month
Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Managed Agents bills runtime only while the agent is working, so short bursts of code with long gaps between them cost little runtime.	186th of 345, with limits $2,000 a month	163rd of 345, with limits $756 a month
Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Your users can't open what their Managed Agents sandbox is running, because it has no inbound ports; bochka gives each sandbox HTTPS URLs on up to 50 ports.	Cheapest of 361 $209 a month	231st of 361, with limits $823 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	252nd of 360, with limits $3,978 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	244th of 364, with limits $320 a month
CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month CI jobs that build Android apps or VM images need /dev/kvm and a full Docker, which bochka has and Managed Agents does not document.	4th cheapest of 377 $86 a month	227th of 377, with limits $96 a month
24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB A dev box needs SSH and a disk that lasts; Managed Agents has no SSH and reaches the sandbox only through the session API.	18th cheapest of 368 $26 a month	224th of 368, with limits $48 a month
Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Managed Agents ships headless Chromium with Playwright but no desktop or computer-use tool; bochka streams a full Linux desktop at 60fps.	6th cheapest of 369 $200 a month	127th of 369, with limits $288 a month
Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month Both can script Chromium with Playwright; bochka also shows the browser on a live desktop you can watch or take over.	7th cheapest of 367 $200 a month	126th of 367, with limits $288 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	127th of 346, with limits $390 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	155th of 338, with limits $4,798 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	179th of 360, with limits $160 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	Claude Managed Agents
What it is	A sandbox platform: each agent gets its own full Linux machine, with coding agents built in	A hosted Claude agent: Anthropic's harness runs Claude in a sandbox and you send it events
Isolation	Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts	Isolated Linux container per session
Machine size	Up to 16 vCPU / 32 GB on the $100 plan and above	Up to 8 GB memory and 10 GB disk; vCPU not published; not configurable
Docker, systemd and /dev/kvm	Root, Docker, systemd and nested virtualization for Android emulators and VMs	Docker listed as limited availability; root, systemd and /dev/kvm not documented
What persists	The whole disk, saved every minute and on stop, for as long as you keep the sandbox	Sandbox is checkpointed when a session goes idle; state is kept only 30 days after the sandbox is created
Fork	Fork a running sandbox with its disk	No fork or clone of a session or sandbox
Wake on next message	No; a stopped sandbox is started by API	Yes; an idle session resumes from its checkpoint on the next message, and idle time is not billed
Inbound access	HTTPS URLs for up to 50 ports, token-protected by default; raw TCP; SSH	No inbound ports or SSH; the sandbox is reached only through the sessions and events API
Browser and desktop	Full Linux desktop streamed at 60fps, with Chromium and computer use	Headless Chromium with Playwright; no desktop or computer-use tool
Agents	Claude Code, Codex and five more preinstalled, driven from outside with a prompt, steer and events API	Anthropic's harness with Claude models, SSE events, steering, interrupts, multiagent threads, memory stores and webhooks
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	Vault secrets live in the sandbox as placeholders and are swapped in at egress, per host
Outbound rules	Open internet, no provider-side allowlist	Unrestricted, or limited to listed hosts plus optional package registries and MCP servers
Self-hosting	No	Self-hosted sandboxes run tools on your own workers, with guides for several sandbox providers
Where data lives	EU only (Germany, Finland, France), with zero data retention	US workspace geo only; sessions are stored server-side and not eligible for Zero Data Retention
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	Claude Managed Agents
Full VM (own kernel)	Yes	No
Docker inside	Yes	Not published
Nested virtualization	Yes	No
Root	Yes	Not published
Systemd	Yes	Not published
Desktop GUI	Yes	No
Browser	Yes	Yes
Computer-use API	Yes	No
SSH	Yes	No
Persistent disk	Yes	Yes
Pause/resume	Yes	Yes
Automatic snapshots	Yes	Yes
Fork/clone	Yes	No
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	No
Raw TCP inbound	Yes	No
Public IPv4	No	No
Egress allowlist	No	Yes
Secret proxy	Yes	Yes
Secret proxy for any API	Yes	Yes
Preinstalled agents	Yes	Yes
Agent harness API	Yes	Yes
MCP server	No	Yes
GPU	No	No
Windows	No	No
Arm64	No	No
Self-hosting	No	Yes
BYOC	No	Yes
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	No
Zero data retention	Yes	No

Performance

bochka26.69 runs/s

Claude Managed AgentsNot measured

There is no comparable measurement: Managed Agents was not part of the hpc-sandbox-benchmarks run, and Anthropic does not publish how many vCPUs a cloud sandbox gets. bochka did 26.69 runs per second on the Node.js build-and-test loop at 4 vCPU / 8 GB. On Managed Agents, most of a session's time is often spent waiting on the model rather than on the CPU.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Claude Managed Agents 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	Claude Managed Agents, 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	$13 · Claude API (pay-as-you-go, usage tiers) · with limits
One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month	$26 · $20/mo	$53 · Claude API (pay-as-you-go, usage tiers) · 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	$634 · Claude API (pay-as-you-go, usage tiers) · 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	$360 · Claude API (pay-as-you-go, usage tiers) · 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	$796 · Claude API (pay-as-you-go, usage tiers) · with limits

bochka: Plans start at $20 a month, and the whole fee is usage credit. You pay per second while a sandbox runs, and a stopped sandbox costs nothing and keeps its disk; model tokens for the built-in agents are paid to your model provider and are not in the table.

Claude Managed Agents: Managed Agents bills model tokens at API list rates plus a flat session runtime fee, measured to the millisecond only while the session is running, including time spent waiting on the model. The table shows runtime only; tokens come on top and usually cost more than the runtime.

Limits in those scenarios

A Managed Agents sandbox has up to 8 GB of memory and 10 GB of disk, its vCPU count is not published, and Docker is only in limited availability, which is why no scenario on this page counts as a clean fit for it. Anthropic publishes no concurrent-session cap, only 300 creates per minute per organization and usage-tier spend limits that pause all API use when reached. bochka has no session cap, but at 1,000 short jobs at once the $2,000 plan queues starts above 90 per minute and allows 2,400 starts a day.

Scenario	bochka	Claude Managed Agents
One coding agent, workdays	None that bite	adjacent product, not a like-for-like sandbox machine size not published different kind of product: agent-platform machine size per session not published
One agent that never stops	None that bite	adjacent product, not a like-for-like sandbox machine size not published different kind of product: agent-platform machine size per session not published
50 agents in parallel	None that bite	adjacent product, not a like-for-like sandbox not confirmed: Docker inside machine size not published different kind of product: agent-platform
Sandboxes for your users	None that bite	adjacent product, not a like-for-like sandbox machine size not published different kind of product: agent-platform not confirmed: how many machines can run at once (you need 300; the typical published limit is 50)
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)	adjacent product, not a like-for-like sandbox machine size not published different kind of product: agent-platform not confirmed: how many machines can run at once (you need 1,000; the typical published limit is 50)

Which one should you use?

Use Claude Managed Agents when

you want Claude itself to do long, mostly text and code tasks with no harness to build, idle time that costs nothing, and the option to run tools on your own workers.

No agent loop to build. Teams that want Claude doing long tasks get the harness, caching, compaction, memory and webhooks ready to use.
Idle sessions sleep for free and wake on the next message. Agents that wait a long time between messages pay runtime only while working, and resume on the next message.
Host allowlists and your own workers. Teams with strict network rules can limit sandboxes to listed hosts, or run every tool call on their own infrastructure.

Use bochka when

your agents need a real machine: Docker, emulators, a desktop and browser, ports you can open, a disk that lasts, and the choice of Claude Code, Codex or other agents.

A whole machine, not a small container. Builds that need Docker, a database cluster or an Android emulator stall in a small container. And an agent that comes back after a month on Managed Agents finds a fresh sandbox, so all its setup runs again.
Open what the agent builds. When the agent starts a dev server, you and your users want to click a link and try it. On Managed Agents you would have to copy the code somewhere else to run it.
Any agent, driven from outside. Managed Agents runs only Anthropic's harness with Claude models. If your team already works in Claude Code, or wants to try Codex on the same task, that has to happen somewhere else.

Questions people ask

Is bochka cheaper than Claude Managed Agents?

For runtime alone, bochka costs less in three of the five scenarios on this page: an agent that never stops, 50 agents in parallel, and sandboxes for your users. For one agent on workdays and for 1,000 short jobs, Managed Agents runtime costs less because it bills only while the agent is working. Both tables leave out model tokens, which you pay either way.

Can I run Claude Code on Claude Managed Agents?

No. Managed Agents runs Anthropic's own harness with Claude models, and its branding rules say partner products must not appear to be Claude Code. On bochka, Claude Code, Codex and five other agents come preinstalled in the VM.

Can I open a web app my agent built on Claude Managed Agents?

Not directly. The docs describe only outbound network controls, and the sandbox is reached through the sessions API. bochka gives each sandbox HTTPS URLs on up to 50 ports, raw TCP and SSH.

How long do Claude Managed Agents sandboxes keep their files?

An idle session's sandbox is checkpointed and resumes on the next message, but sandbox state is kept only 30 days after the sandbox is created, and activity does not extend it. bochka keeps the whole disk, saved every minute, until you delete the sandbox.

Can I use Claude Managed Agents with a bochka VM?

A self-hosted environment worker needs only a Linux host with /bin/bash and outbound HTTPS, which a bochka VM has. Anthropic does not list bochka among its platform guides, so you install and run the ant worker yourself.

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