GitHub Codespaces vs Agent 37

GitHub Codespaces is a dev environment for people: a Linux dev container opened from a GitHub repository in VS Code or the browser, which stops after 12 hours at most. Agent 37 hosts AI agents: one always-on gVisor sandbox per user behind one API, which can sleep when idle and wake with its memory kept.

The companies and products

GitHub Codespaces

GitHub Codespaces is GitHub's cloud development environment: a ready-to-code Linux machine for a repository, opened in the browser, VS Code or the GitHub CLI.

Codespaces is a different kind of product from bochka: a dev environment for people working on a GitHub repository, not a sandbox API for agents. Each codespace is a Docker container built from the repo's dev container config, running on its own newly built VM on Azure, with 2 to 32 cores. It is built for developers and teams who want every contributor to start from the same setup, with prebuilds, port forwarding and GitHub billing.

Isolation: A Docker container on its own newly built VM, with its own virtual network
Operating system: Linux only; Windows and macOS are not supported for the remote container
Machine sizes: 2, 4, 8, 16 or 32 cores, with 4 GB of memory per core
Lifetime: Stops after 30 idle minutes by default (5 to 240), and after 12 hours at most even while in use
Regions: US East, US West, Europe West and Southeast Asia
Free usage: Personal Free accounts get 120 core-hours and 15 GB-month a month, Pro 180 and 20; organizations get no free quota
Compliance: GitHub Enterprise Cloud lists SOC 1, SOC 2 and SOC 3 reports and ISO 27001

Agent 37

Agent 37 hosts AI agents, one isolated, always-on sandbox per user, behind one API.

You create an instance from a template (Hermes, OpenClaw, Claude Code, Codex, OpenCode, Pi, Grok or your own Docker image) and talk to it through a chat API, files, exec, SSH and HTTPS URLs. The default instance is a container enforced by gVisor; a VM type gives a full Linux machine with its own kernel for Docker, Kubernetes and Android emulators. It is built for companies that give each of their customers an agent.

Isolation: gVisor by default; opt-in full VM type with its own kernel and KVM
Where it runs: Dedicated servers in the United States
Billing: Prepaid wallet, metered per minute, no plan fee or seat charge
Instance types: Default (shared), Performance (dedicated cores), VM
Compliance: SOC 2 Type I, Type II in progress
Company: Y Combinator company based in New York

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. “274th cheapest of 367” means 273 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	GitHub Codespaces: price rank	Agent 37: price rank	bochka: price rank
Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Codespaces gives an agent Docker in a dev container but stops it at 12 hours; Agent 37 needs its VM type for Docker, and bochka has Docker, systemd and /dev/kvm on every machine.	274th of 367, with limits $4,618 a month	2nd cheapest of 367 $551 a month	3rd cheapest of 367 $634 a month
Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Agent 37 is built for this, with one assistant per user that sleeps when idle and wakes with its memory, while every codespace stops after 12 hours at most.	309th of 366, with limits $6,970 a month	Cheapest of 366 $279 a month	5th cheapest of 366 $657 a month
Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Agent 37's small default sandboxes suit short code runs, while each codespace builds a dev container on a new VM, and bochka's start limits make big bursts a poor fit.	288th of 345, with limits $619 a month	Cheapest of 345 $22 a month	186th of 345, with limits $2,000 a month
Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month GitHub does not publish how many codespaces can run at once, and Agent 37 caps self-serve accounts at 200 instances counting sleeping ones; bochka plans go up to 2,000.	281st of 361, with limits $1,964 a month	30th of 361, with limits $79 a month	Cheapest of 361 $209 a month
RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month	282nd of 360, with limits $9,489 a month	6th of 360, with limits $363 a month	34th of 360, with limits $2,000 a month
Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month	295th of 364, with limits $1,519 a month	90th of 364, with limits $58 a month	10th cheapest of 364 $200 a month
CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Codespaces starts each run from a repo or a prebuild built by GitHub Actions, and Agent 37 restores backups only in place; bochka can fork a warm machine.	314th of 377, with limits $909 a month	2nd cheapest of 377 $84 a month	4th cheapest of 377 $86 a month
24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Codespaces is a polished VS Code box per repo but stops after 12 hours; Agent 37 and bochka both keep a machine and its disk around the clock.	130th of 368, with limits $270 a month	3rd cheapest of 368 $12 a month	18th cheapest of 368 $26 a month
Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month A codespace has an editor and a terminal but no graphical desktop; Agent 37 adds a noVNC desktop through a template, and bochka streams a 60fps desktop with Chromium.	267th of 369, with limits $1,367 a month	Cheapest of 369 $53 a month	6th cheapest of 369 $200 a month
Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month	275th of 367, with limits $678 a month	48th of 367, with limits $29 a month	7th cheapest of 367 $200 a month
Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month	310th of 346, with limits $922 a month	Cheapest of 346 $32 a month	6th cheapest of 346 $90 a month
Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month	280th of 338, with limits $18,637 a month	36th of 338, with limits $648 a month	71st of 338, with limits $1,800 a month
Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month Neither Codespaces nor Agent 37 can copy a running machine, so every attempt starts over; bochka can fork one prepared machine into many while it runs.	267th of 360, with limits $445 a month	Cheapest of 360 $43 a month	2nd cheapest of 360 $200 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	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	Can't run it

Key features

Feature	GitHub Codespaces	Agent 37	bochka
Isolation	A Docker container on its own newly built VM, with its own virtual network	gVisor sandbox by default; full VM with its own kernel as an opt-in type	Every sandbox is a full VM with its own kernel
How long it runs	Stops after 30 idle minutes by default (up to 240) and after 12 hours at most, even while in use	No time limit; runs until you stop it or it sleeps on idle	Runs until you stop it; no session cap on paid plans
Docker inside	Docker and sudo inside the dev container	Only on the VM type, at 2.5x the default compute rate; the default gVisor type cannot run a Docker daemon	Docker, root and systemd in every sandbox
Nested virtualization	/dev/kvm and nested virtualization are not documented	/dev/kvm on the VM type, with a ready Android template	/dev/kvm in every sandbox, Android emulator works
Sleep when idle	Stops on idle; files are kept, running processes end	Opt-in auto-sleep after a quiet period, memory checkpointed, wakes on the next request	No idle timer; disk kept on stop, a resume reboots the machine
Fork and snapshots	No copy or snapshot endpoint; new codespaces start from a repo, branch, pull request or prebuild	No fork; daily backups plus one on-demand backup, restored in place only	Fork a running sandbox; snapshots every minute; named snapshots as shareable templates
Starting point	Dev container config in the repository, with prebuilds built by GitHub Actions	Your own linux/amd64 Docker image as a template, or a ready agent template	Named snapshots (templates), shareable by code; no Dockerfile builds
Coding agents	GitHub Copilot as a VS Code extension; other agents you install and run yourself	Hermes, OpenClaw, Claude Code, Codex, Grok, OpenCode, Pi, with a chat API per instance	Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API
Keeping keys out of the sandbox	Secrets are copied into the codespace as environment variables	Managed LLM, search and Composio through a per-instance token; your own keys go in as env vars	AI credentials proxy for model providers and any HTTPS API you register; switch a credential to another account when credits run out
Desktop and computer use	VS Code in the browser or on your machine, JupyterLab and a terminal; no graphical desktop	noVNC desktop through a template build or the Ubuntu VM template	60fps desktop and Chromium in every sandbox, computer use, embeddable stream
Reaching the sandbox	A URL per forwarded port, private, organization or public; SSH and CLI tunnels; no public IP	HTTPS URL per port, signed URLs, custom domains; no public IP, SSH through a WebSocket tunnel	HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4
CPU	2, 4, 8, 16 or 32 cores, with 4 GB of memory per core	Shared by default; Performance type on dedicated cores at 4x the compute rate	Shared vCPUs on dedicated AMD Ryzen 9 9950X hosts
How many at once	Limits exist but GitHub does not publish them	1 instance before a top-up, then 10, 50 or 200 by lifetime top-ups, counting sleeping and stopped ones	100 to 2,000 sandboxes by plan, plus a cap on starts per minute
Regions	US East, US West, Europe West and Southeast Asia	Dedicated servers in the United States	EU only: Germany, Finland, France, with EU data residency

Every feature Battleships tracks (43)

Feature	GitHub Codespaces	Agent 37	bochka
Full VM (own kernel)	Yes	No	Yes
Docker inside	Yes	No	Yes
Nested virtualization	No	No	Yes
Root	Yes	Yes	Yes
Systemd	Not published	Not published	Yes
Desktop GUI	No	Yes	Yes
Browser	No	Yes	Yes
Computer-use API	No	Yes	Yes
SSH	Yes	Yes	Yes
Persistent disk	Yes	Yes	Yes
Pause/resume	Yes	Yes	Yes
Automatic snapshots	No	Yes	Yes
Fork/clone	No	Yes	Yes
Live fork (no pause)	No	Yes	Yes
Memory fork	No	No	No
Custom image (Docker or snapshot)	Yes	Yes	Yes
≥24 h sessions	No	Yes	Yes
Idle auto-stop	Yes	Yes	No
Wake on request	No	Yes	No
HTTPS preview URLs	Yes	Yes	Yes
Raw TCP inbound	No	Not published	Yes
Public IPv4	No	No	No
Egress allowlist	No	No	No
Secret proxy	No	Yes	Yes
Secret proxy for any API	No	No	Yes
Preinstalled agents	No	Yes	Yes
Agent harness API	No	Yes	Yes
MCP server	No	No	No
GPU	No	No	No
Windows	No	No	No
Arm64	No	No	No
Self-hosting	No	No	No
BYOC	No	No	No
Open source	No	No	No
Scoped API keys	Not published	Not published	Yes
Spend limits	Yes	Yes	Yes
Usage per sandbox (API)	No	Yes	Yes
Audit logs	Yes	Not published	No
SSO	Yes	No	No
SOC 2	Yes	Yes	No
HIPAA	No	No	No
EU data residency	Yes	No	Yes
Zero data retention	Not published	Not published	Yes

Performance

GitHub CodespacesNot measured

Agent 37Not measured

bochka26.69 runs/s

There is no comparable measurement for GitHub Codespaces or Agent 37: neither was in the public benchmark runs we use. bochka did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB in StarSling's run of 2026-09-29, on dedicated AMD Ryzen 9 9950X hosts. Agent 37 says its default instances share host resources and sells dedicated cores on its Performance type.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. GitHub Codespaces and Agent 37 have 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	GitHub Codespaces, a month	Agent 37, a month	bochka, 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	$51 · Personal account: GitHub Free	$6 · Wallet (any top-up)	$20 · $20/mo
One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month	$270 · Personal account: GitHub Free · with limits	$12 · Wallet (any top-up)	$26 · $20/mo
50 agents in parallel 50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside	$3,019 · Personal account: GitHub Free	$329 · Wallet ($100 lifetime top-ups)	$317 · $20/mo × 3 seats
Sandboxes for your users 2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once	$850 · Personal account: GitHub Free · with limits	$33 · Wallet ($250 lifetime top-ups) · with limits	$200 · $100/mo × 2 seats
1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month	$1,889 · Personal account: GitHub Free · with limits	$73 · Wallet ($250 lifetime top-ups) · with limits	$2,000 · $2000/mo · with limits

GitHub Codespaces: Codespaces bills $0.09 per core-hour while a codespace is active, plus $0.07 per GB-month of storage while it exists, stopped too. There is no 4-core machine with 8 GB, so 4 vCPU / 8 GB jobs are priced on the 4-core 16 GB machine; personal accounts get a free monthly quota and organizations do not.

Agent 37: Agent 37 uses a prepaid wallet with no plan fee, metered per minute: $0.80 per vCPU and $0.70 per GB of RAM per month while running, and $0.09 per GB of disk per month even while stopped or asleep. Performance compute costs 4x and VM compute 2.5x the default rate.

Agent 37: 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.

Limits in those scenarios

Every codespace stops after 12 hours at most, so it does not fit the always-on scenario, and GitHub does not publish how many can run at once, so 300 user sandboxes and 1,000 short jobs are not confirmed. Agent 37 caps self-serve accounts at 200 instances, counting sleeping and stopped ones, so those two scenarios do not fit there either. The 50-agent scenario needs Docker, so Agent 37 is priced on its VM type at 2.5x the compute rate. bochka fits every scenario but the 1,000-job burst, where its cap on starts per day and per hour stops it.

Scenario	GitHub Codespaces	Agent 37	bochka
One coding agent, workdays	None that bite	egress price not published	None that bite
One agent that never stops	Personal account: GitHub Free: sessions capped at 12 h	egress price not published	None that bite
50 agents in parallel	how many machines can run at once: limit not published, probably fine at 50 (the typical published limit is 50)	egress price not published	None that bite
Sandboxes for your users	not confirmed: how many machines can run at once (you need 300; the typical published limit is 50)	max 200 concurrent egress price not published	None that bite
1,000 short jobs at once	not confirmed: how many machines can run at once (you need 1,000; the typical published limit is 50)	max 200 concurrent shared vCPUs on this tier: heavy, sustained CPU use can be slowed by other tenants egress price not published	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)

Which one should you use?

Use GitHub Codespaces when

your developers want a ready VS Code environment for a GitHub repository, opened from a branch or pull request, in the US, EU or Asia.

Built into GitHub. Useful if your team lives on GitHub: open a codespace from any repo, branch or pull request, with the dev container already set up.
Free hours for personal accounts. Useful for occasional personal work: a GitHub Free account includes 120 core-hours and 15 GB-month of storage each month.
Regions in the US, EU and Asia. Useful if your developers are spread out: codespaces run in US East, US West, Europe West and Southeast Asia.

Use Agent 37 when

you host one mostly idle assistant per customer in the US that should sleep with its memory kept and wake on the next message.

Sleeps when idle and keeps its memory. Useful for assistants that sit idle most of the day: an instance sleeps, keeps its memory and wakes on the next message.
Built to host one agent per customer. Useful if you give each customer an agent: templates for Hermes, OpenClaw, Claude Code and others, with a chat API per instance.
A small bill for idle agents. Useful when budget matters: a prepaid wallet billed per minute with no plan fee, and the lowest bill of the two in every scenario we priced.

Where bochka fits

Codespaces is made for people editing a repository, and Agent 37 for hosting mostly idle assistants. If your code starts agents that do real work for hours, like installs, builds, Docker and emulators, look at bochka: every sandbox is a full VM that keeps running, keeps its work and can be forked while it runs.

A full VM with Docker and /dev/kvm by default

Codespaces gives Docker in a dev container but documents no /dev/kvm. Agent 37's default gVisor type cannot run Docker, and its VM type costs 2.5 times as much and has no init system, so services do not start on their own.

Work that keeps running and can be forked

A codespace stops at 12 hours even while in use, so you need a restart script, and its processes end. Neither can copy a running machine, so every parallel try rebuilds its state from the start.

Keys stay outside, for any API

Codespaces copies secrets in as environment variables, and Agent 37 keeps only its own managed models, search and Composio outside. Your own keys sit inside a machine the agent controls, where one prompt injection can leak them.

Use bochka when your code starts and drives agents that work for hours: full VMs with Docker, systemd and /dev/kvm, a 60fps desktop, open ports, forks of running machines and keys kept outside for any API. Compare it one to one: bochka vs GitHub Codespaces and bochka vs Agent 37.

Questions people ask

Is GitHub Codespaces cheaper than Agent 37?

No, in every scenario we priced Agent 37 is lower. One coding agent on workdays is $6 a month on Agent 37 against $51 on Codespaces, and 50 agents in parallel are $329 against $3,019. bochka comes in at $20 and $317 for those two.

Which is faster, GitHub Codespaces or Agent 37?

We have no comparable CPU measurement for either one. bochka did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB on the StarSling benchmark. Agent 37 sells dedicated cores on its Performance type for steady speed.

Can I run an AI agent 24/7 on GitHub Codespaces or Agent 37?

Not on Codespaces without restarts: a codespace stops after 12 hours at most, even while in use. Agent 37 has no time limit and can sleep on idle. A bochka sandbox runs until you stop it, with no session cap on paid plans.

Can GitHub Codespaces and Agent 37 run Docker?

Codespaces runs Docker inside the dev container, and you have sudo. Agent 37 runs Docker only on its VM type, at 2.5x the default compute rate. Every bochka sandbox has Docker, systemd and /dev/kvm.

Where do GitHub Codespaces and Agent 37 run?

Codespaces runs in US East, US West, Europe West and Southeast Asia. Agent 37 runs on dedicated servers in the United States. bochka runs only in the EU, in Germany, Finland and France, with EU data residency.

Try bochka

Install the CLI, then run bochka new.

Product preview