OpenAI Containers vs Agent 37

OpenAI Containers are short-lived boxes inside OpenAI's model API, where the model runs code and the box expires after 20 idle minutes. Agent 37 gives each of your users a persistent agent sandbox that sleeps when idle and keeps its disk and memory.

The companies and products

OpenAI Containers

OpenAI Containers are the sandboxes OpenAI runs for its own model tools: Code Interpreter, Hosted Shell and the Agents API's hosted sandbox.

This is a different kind of product from boat: a tool inside OpenAI's model API, not a machine you rent. You create a container through the API or let a Responses call make one, and the model runs Python or shell commands in it on a fixed Debian 12 image without sudo, until it expires after 20 minutes without use. It is built for apps that already call OpenAI models and want them to run code, read files and make charts without running any servers.

Isolation: Each container is described as a fully sandboxed virtual machine; the hypervisor is not named
Runtime: Debian 12 with Python 3.11, Node.js 22.16, Java 17, PHP 8.2, Ruby 3.1 and Go 1.23; commands do not run with sudo
Sizes: Picked by memory: 1 GB (default), 4 GB, 16 GB or 64 GB; vCPU per size not published for Responses containers
Lifetime: A container expires after 20 minutes without use, and its data is discarded
Billing: By the minute with a 5-minute minimum per session, on top of model tokens
Regions: Code Interpreter processing in the United States or Europe; the Agents API supports data residency only in the US
Compliance: API Platform audited for SOC 2 Type 2; HIPAA BAA on request

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. “356th cheapest of 367” means 355 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	OpenAI Containers: price rank	Agent 37: price rank	boat: price rank
Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A coding agent needs root, Docker and a disk that lasts for days; OpenAI Containers have none of these, Agent 37 needs its VM type for Docker, and boat has all three on every sandbox.	356th of 367, with limits $17,371 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 one always-on assistant per user that sleeps and wakes with its memory; an OpenAI container is deleted after 20 idle minutes.	341st of 366, with limits $13,140 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 Running model-written Python inside one API call is what OpenAI Containers are built for; Agent 37's small default sandboxes also suit short code runs and cost far less in the table.	194th of 345, with limits $6,000 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 Agent 37 gives each user a sandbox with HTTPS URLs but tops out at 200 instances self-serve; OpenAI containers have no inbound access, and boat plans go up to 2,000 at once.	361st of 361, with limits $3,780 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	191st of 360, with limits $30,000 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	360th of 364, with limits $5,760 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 CI jobs need Docker, root and open internet; OpenAI documents no Docker or root and starts Code Interpreter with no internet, while Agent 37 and boat both have open internet.	345th of 377, with limits $1,728 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 Agent 37 and boat both keep a machine and its disk all month with SSH; OpenAI containers have no SSH and no persistent disk.	362nd of 368, with limits $1,051 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 OpenAI's hosted browser runs inside the Agents API beta, Agent 37 adds a noVNC desktop through a template, and boat streams a 60fps desktop in every sandbox.	351st of 369, with limits $5,280 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	328th of 367, with limits $1,440 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	328th of 346, with limits $1,755 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	311th of 338, with limits $35,100 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 boat can fork a running sandbox to try many paths at once; neither OpenAI Containers nor Agent 37 has a fork of a running workspace.	308th of 360, with limits $870 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	OpenAI Containers	Agent 37	boat
What it is	A sandbox tool the model uses inside OpenAI's API	One persistent agent sandbox per user, behind one API	A cloud VM you create, keep, stop and reach from outside
Isolation	A fully sandboxed virtual machine; hypervisor not named	gVisor by default; full VM as an opt-in instance type	Every sandbox is a full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
Root and Docker	Commands do not run with sudo; no Docker documented	Root over SSH; Docker only on the VM type, at 2.5x the default compute rate	Root, systemd and Docker in every sandbox
Nested virtualization	No /dev/kvm documented	/dev/kvm on the VM type, with an Android template	/dev/kvm in every sandbox, Android emulator works
Machine size	Picked by memory: 1, 4, 16 or 64 GB; vCPU not published	Shapes from 2 vCPU / 4 GB to 16 vCPU / 32 GB; Performance type on dedicated cores	4 vCPU / 8 GB by default, up to 16 vCPU / 32 GB
Files after the session	Data is discarded when the container expires	Disk kept; daily backups plus one on-demand backup, restored in place	Disk saved every minute and on stop; a stopped sandbox costs nothing and comes back
Lifetime and idle	Expires after 20 minutes without use	Always on until stopped; opt-in sleep on idle that keeps memory and wakes on the next request	No session cap on paid plans; you set the time to live, resume reboots with the disk kept
Fork	No fork	No fork of a running instance; backups restore only onto the same instance	Fork a running sandbox with its disk; named snapshots as templates
Custom image	Fixed Debian 12 image; you upload files or skills	Your own linux/amd64 Docker image as a template	Templates from named snapshots, no Dockerfile builds
Inbound access	None; reached only through the API	HTTPS URL per port, custom domains, SSH through a WebSocket tunnel; no public IP	SSH, HTTPS URLs for up to 50 ports, raw TCP
Outbound internet	Code Interpreter and Hosted Shell: off until an admin sets a domain allowlist	Open internet	Open internet on every plan
Keys kept out of the sandbox	domain_secrets: a sidecar adds the real header for approved domains	Managed models, search and Composio through a per-instance token; your own keys go in as env vars	Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out
Coding agents and desktop	Agents API (beta) drives a managed Codex harness and a hosted browser	Hermes, OpenClaw, Claude Code, Codex and more with a chat API; noVNC desktop through a template	Claude Code, Codex, OpenCode and more with a prompt, steer and events API; 60fps desktop in every sandbox
Regions	US or EU processing for Code Interpreter; Agents API in the US	United States	EU only (Germany, Finland, France), EU data residency

Every feature Battleships tracks (43)

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

Performance

OpenAI ContainersNot measured

Agent 37Not measured

boat26.69 runs/s

There is no comparable measurement for either: OpenAI Containers and Agent 37 are not in the public CPU benchmark, and OpenAI does not publish the vCPU behind each container size. boat did 26.69 runs per second on the hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU / 8 GB. 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. OpenAI Containers 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	OpenAI Containers, a month	Agent 37, a month	boat, 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	$228 · API pay as you go · with limits	$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	$1,051 · API pay as you go · 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	$11,405 · API pay as you go · with limits	$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	$1,620 · API pay as you go · 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	$6,000 · API pay as you go · with limits	$73 · Wallet ($250 lifetime top-ups) · with limits	$2,000 · $2000/mo · with limits

OpenAI Containers: OpenAI bills each container session by the minute for the memory size you pick, with a 5-minute minimum, and model tokens are billed separately on top. There is no plan fee.

Agent 37: 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, plus $0.09 per GB of disk per month even while stopped or asleep. Performance compute costs 4x and VM compute 2.5x.

Agent 37: A monthly plan from $20, all of it usage credit spent per second on running sandboxes. A stopped sandbox costs nothing and keeps its disk.

Limits in those scenarios

OpenAI Containers fit none of the five scenarios as written: no persistent disk, no Docker, no internet in Code Interpreter and Hosted Shell until an admin allows domains, and no published vCPU or concurrency limit. Agent 37 fits the single-agent and fleet scenarios, with the fleet priced on its VM type because it needs Docker, but its cap of 200 instances at $250 of top-ups, counting sleeping and stopped ones, rules out 300 user sandboxes and the 1,000-job burst. boat's limit in the burst is starts per minute, so those starts are queued.

Scenario	OpenAI Containers	Agent 37	boat
One coding agent, workdays	no open internet: sandboxes reach only an allowlist (domains your org admin adds to the allow list) missing persistent disk machine size not published different kind of product: code interpreter	egress price not published	None that bite
One agent that never stops	no open internet: sandboxes reach only an allowlist (domains your org admin adds to the allow list) missing persistent disk machine size not published different kind of product: code interpreter	egress price not published	None that bite
50 agents in parallel	no open internet: sandboxes reach only an allowlist (domains your org admin adds to the allow list) missing persistent disk, Docker inside machine size not published different kind of product: code interpreter	egress price not published	None that bite
Sandboxes for your users	no open internet: sandboxes reach only an allowlist (domains your org admin adds to the allow list) machine size not published different kind of product: code interpreter 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	no open internet: sandboxes reach only an allowlist (domains your org admin adds to the allow list) machine size not published different kind of product: code interpreter 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 OpenAI Containers when

your app already calls OpenAI models and the model only needs to run short pieces of code or shell commands on files you give it.

Built into the model call. If your app already calls OpenAI models, one tool line in a request gives the model a place to run code.
Nothing to run or clean up. OpenAI creates the container, bills it by the minute with your tokens, and it expires on its own when idle.
Locked-down network by default. Code Interpreter and Hosted Shell reach no hosts until an admin allows domains, which suits code that must stay contained.

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.

A persistent machine per user. Each of your customers gets an always-on sandbox with its own disk, SSH and HTTPS URLs, created from an agent template.
Sleeps and keeps its memory. For assistants idle most of the day, an instance sleeps, saves its memory and wakes on the next message.
Bring your own Docker image. Templates build from your own Docker image, and a Performance type runs on dedicated cores for steady speed.

Where boat fits

OpenAI Containers suit short code runs inside a model call, and Agent 37 suits mostly idle assistants that sleep. If your agents do real work for hours, with installs, Docker, builds and apps people need to open, boat gives every one a full VM it keeps, in the EU.

A full VM with Docker in every sandbox

An OpenAI container runs without sudo and has no Docker. On Agent 37 the default gVisor type cannot run Docker; the VM type can, but bills compute at 2.5x and has no systemd, so services do not start on their own.

Fork a machine while it runs

Neither has a fork. An OpenAI container's data is gone when it expires, and Agent 37 backups restore only onto the same instance, so each new try rebuilds a half-done task from the start.

Keys outside, apps reachable

OpenAI containers accept no inbound connections, so nobody can open the app an agent built. On Agent 37 your own API keys sit as env vars inside the machine; boat's proxy keeps them outside for any HTTPS API.

Use boat when your agents do real work for hours: installs, builds, Docker and emulators on a full VM, with forks of running machines, ports people can open and keys kept outside for any API. Compare it one to one: boat vs OpenAI Containers and boat vs Agent 37.

Questions people ask

Is OpenAI Containers cheaper than Agent 37?

No, not in the scenarios on this page: one coding agent on workdays comes to $228 a month on OpenAI and $6 on Agent 37, and one agent running all month to $1,051 versus $12. OpenAI also bills model tokens on top, and its containers do not fully fit these scenarios.

Which is faster, OpenAI Containers or Agent 37?

Neither is in the public CPU benchmark, so there is no fair comparison. OpenAI does not publish the vCPU per container size, and Agent 37 sells dedicated cores on its Performance type. boat scored 26.69 runs per second on that benchmark.

Can OpenAI Containers or Agent 37 run Docker?

OpenAI documents no Docker, and hosted shell commands do not run with sudo. Agent 37 runs Docker only on its VM type, at 2.5x the default compute rate. boat runs Docker in every sandbox.

Do they keep files between sessions?

An OpenAI container expires after 20 minutes without use and its data is discarded. Agent 37 keeps the disk and can sleep with memory kept. boat saves the disk every minute and when the sandbox stops.

Can I open a web app running inside them?

Not on OpenAI: containers have no ports, URLs or SSH. Agent 37 gives an HTTPS URL per port and SSH through a WebSocket tunnel. boat gives HTTPS URLs for up to 50 ports, raw TCP and SSH.

Where do they run?

OpenAI processes Code Interpreter data in the US or Europe, and the Agents API in the US. Agent 37 runs on dedicated servers in the United States. boat runs only in the EU, in Germany, Finland and France.

Try boat

Install the CLI, then run boat new.

7-day free trial