OpenAI Containers vs Islo

OpenAI Containers are a tool inside OpenAI's model API: the model runs short pieces of code in a box that is deleted after 20 minutes without use. Islo gives each coding agent its own microVM with sudo, Docker and a disk that stays, with a gateway that controls what it can reach.

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 bochka: 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

Islo

Islo is an agent sandbox from Incredibuild: each coding agent gets its own microVM, with network rules and credentials handled at a gateway.

An Islo sandbox is a Linux microVM with its own kernel, built from a container image, with Claude Code, Cursor CLI and Codex preinstalled. Outbound traffic can pass through a gateway that allows or blocks hosts, injects credentials and logs every request. It is built for teams that run coding agents, evals and RL rollouts in the background, on Islo's cloud or in their own cloud account.

Maker: Incredibuild, a build acceleration company based in Tel Aviv
Launched: May 4, 2026
Isolation: A microVM with its own kernel for each agent computer
Regions: US West on Islo's cloud; BYOC puts the control plane and sandboxes in your own cloud account
Billing: CPU per core, memory per GB and disk per GB while a sandbox runs, billed per hour; signup credits with no card
Compliance: No certification named on its site; compliance documents on request from support

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	Islo: price rank	bochka: price rank
Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A coding agent that keeps a repo for days needs a disk that stays and Docker, which Islo has and an OpenAI container does not; bochka adds /dev/kvm and a desktop for emulators and UI tests.	356th of 367, with limits $17,371 a month	326th of 367, with limits $7,402 a month	3rd cheapest of 367 $634 a month
Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Islo fits an assistant that lives on its machine all month and can pause it with memory between tasks; an OpenAI container is deleted after 20 minutes without use.	341st of 366, with limits $13,140 a month	30th cheapest of 366 $9,928 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 This is what OpenAI Containers are built for: the model writes Python and runs it in the same API call; Islo and bochka are whole machines that also fit, but you run the code yourself.	194th of 345, with limits $6,000 a month	153rd of 345, with limits $503 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 Neither fits 300 user sandboxes at once as published: Islo's Team plan is reported at 50 running sandboxes, and OpenAI publishes no limit at all.	361st of 361, with limits $3,780 a month	171st of 361, with limits $3,160 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	167th of 360, with limits $15,265 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 Islo has a Harbor integration and gateway rules that stop trials from looking up answers; OpenAI containers have no Docker for eval images.	360th of 364, with limits $5,760 a month	198th of 364, with limits $2,456 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: Islo has all three, while OpenAI documents no Docker or sudo and Code Interpreter containers start with no internet.	345th of 377, with limits $1,728 a month	119th cheapest of 377 $1,507 a month	4th cheapest of 377 $86 a month
24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Islo gives you SSH through its CLI and a disk that stays, so it works as a dev box; OpenAI containers have no SSH and no persistent disk.	362nd of 368, with limits $1,051 a month	112th cheapest of 368 $402 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 has a hosted browser for computer use inside the Agents API beta, Islo sandboxes are headless, and bochka streams a 60fps desktop agents can see and click.	351st of 369, with limits $5,280 a month	310th of 369, with limits $2,210 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	113th of 367, with limits $1,105 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	66th cheapest of 346 $748 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	265th of 338, with limits $14,966 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	308th of 360, with limits $870 a month	47th of 360, with limits $746 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	Islo	bochka
What it is	A sandbox tool the model uses inside OpenAI's API	An agent sandbox: a microVM per coding agent	A cloud VM you create, keep, stop and reach from outside
Isolation	A fully sandboxed virtual machine; hypervisor not named	A microVM with its own kernel	Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
Root, Docker and nested virtualization	Commands do not run with sudo; no Docker or /dev/kvm documented	Sudo and Docker inside; no /dev/kvm or nested virtualization documented	Root, systemd, Docker and /dev/kvm on every machine
Files after the session	Data is discarded when the container expires and cannot be recovered	Disk stays; pause keeps files, processes and memory	Disk saved every minute and on stop; a stopped sandbox costs nothing and comes back
Lifetime	Expires after 20 minutes without use; Agents API sandboxes after an hour without activity	No published session cap; optional pause and delete timers set at creation	No session cap on paid plans; you set the time to live
Wake on request	No	Pauses after idle time and resumes on the next exec, file access or webhook	No; a resumed sandbox reboots from disk
Fork	No fork	Snapshot a running sandbox's files, then start new sandboxes from it	Fork a running sandbox into new ones
Custom image	Fixed Debian 12 image; you upload files into each container	Built from a container image you choose	Named snapshots of a set-up machine, reused as templates
Reaching the sandbox	None; only through the API	HTTPS share links that expire after a TTL (24 hours by default); SSH and port-forward through the CLI; no raw TCP	SSH, token-protected HTTPS URLs for up to 50 ports, raw TCP
Outbound internet	Code Interpreter and Hosted Shell: off until an admin sets an org domain allowlist	Open, or gateway rules by host, path and HTTP method with default deny and request logs	Open internet on every plan; no provider-side allowlist
Keeping API keys out	domain_secrets: a sidecar adds the real header for approved domains	Gateway injects stored credentials into outbound requests for any host	Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out
Coding agents	Agents API (beta) drives a managed Codex harness	Claude Code, Cursor CLI and Codex preinstalled; factory lines with events, steer and triggers	Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API
Desktop and computer use	Hosted browser for the Agents API computer_use tool	A browser for testing; no desktop or computer-use API	60fps desktop and Chromium, computer use, embeddable stream
Where it runs	US or EU processing for Code Interpreter; Agents API can drive a sandbox you host	US West, or inside your own cloud account with BYOC	EU only: Germany, Finland, France

Every feature Battleships tracks (43)

Feature	OpenAI Containers	Islo	bochka
Full VM (own kernel)	Yes	Yes	Yes
Docker inside	No	Yes	Yes
Nested virtualization	No	No	Yes
Root	No	Yes	Yes
Systemd	Not published	Not published	Yes
Desktop GUI	Yes	No	Yes
Browser	Yes	Yes	Yes
Computer-use API	Yes	No	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	No	Yes
Public IPv4	No	No	No
Egress allowlist	Yes	Yes	No
Secret proxy	Yes	Yes	Yes
Secret proxy for any API	Yes	Yes	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	Yes	No
Open source	No	No	No
Scoped API keys	Not published	Not published	Yes
Spend limits	Not published	Not published	Yes
Usage per sandbox (API)	No	Yes	Yes
Audit logs	Not published	Yes	No
SSO	Yes	Yes	No
SOC 2	Yes	Not published	No
HIPAA	Yes	Not published	No
EU data residency	Yes	No	Yes
Zero data retention	Not published	Not published	Yes

Performance

OpenAI ContainersNot measured

IsloNot measured

bochka26.69 runs/s

There is no comparable measurement for either: OpenAI Containers and Islo are not in the public CPU benchmark, and OpenAI does not publish how many vCPU each container size gets. 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.

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 Islo 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	Islo, 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	$228 · API pay as you go · with limits	$101 · Free	$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	$420 · Free	$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	$4,863 · Team	$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	$1,382 · Team · 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	$3,053 · Team · 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.

Islo: Islo bills each CPU core and each GB of memory and disk while a sandbox runs, billed per hour, with no plan fee. An idle sandbox keeps billing until a pause-after-idle timer fires, and egress has no published price.

Islo: 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

OpenAI Containers fit none of the five scenarios as written: they have no persistent disk and no Docker, Code Interpreter containers start with no internet, and the vCPU behind each size is not published. Islo fits the three agent scenarios, but its Team plan is reported at 50 running sandboxes, so 300 user sandboxes and 1,000 short jobs at once need a custom contract. OpenAI publishes no limit on containers at once, so those two are unconfirmed there. On bochka the 1,000-job burst is limited by starts per minute and per day, so the $2,000 plan queues it.

Scenario	OpenAI Containers	Islo	bochka
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)	Team: max 50 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)	Team: max 50 concurrent 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. Useful if your app already calls OpenAI models: one tool line in a request gives the model a place to run code.
Nothing to clean up. Useful for short jobs on files you hand over: the container expires on its own after 20 idle minutes.
Locked network by default. Useful if model code must reach nothing: Code Interpreter and Hosted Shell have no outbound access until an admin allows domains.

Use Islo when

your coding agents need a machine they keep, with Docker and a disk that stays, plus per-host egress rules, request logs or pausing with memory.

A machine the agent keeps. Useful for coding agents that work for hours or days: Islo gives sudo, Docker and a disk that stays.
Pause with memory, wake on request. Useful for agents that sit idle between tasks: Islo pauses them with memory kept and resumes on the next request.
Egress rules and request logs. Useful if you must control what agents reach: Islo allows or blocks each host, path and method, and logs every call.

Where bochka fits

If you are choosing between these two, you are probably running agents that write and run code. Look at bochka when those agents need a full VM with /dev/kvm, a desktop they can see, and ports other people can reach, on a disk that is saved every minute.

Emulators and VMs inside

bochka gives root, systemd, Docker and /dev/kvm. OpenAI containers have no sudo or Docker, and Islo has Docker but no /dev/kvm, so Android emulators and tools that start their own VM stop at boot on both.

A desktop agents can see

Every bochka sandbox streams a 60fps desktop with a computer-use API. Islo sandboxes are headless, so you would build a VNC stack into each image, and OpenAI's hosted browser exists only inside the Agents API beta.

Ports you can host on

bochka serves up to 50 ports over token-protected HTTPS, plus raw TCP. OpenAI containers accept no inbound connections at all, and Islo share links are public and expire after 24 hours by default, so you keep re-sharing or run a CLI tunnel.

Use bochka when your agents need a full VM with /dev/kvm, a 60fps desktop for computer use, ports you can host on, and a disk saved every minute, in the EU. Compare it one to one: bochka vs OpenAI Containers and bochka vs Islo.

Questions people ask

Is OpenAI Containers cheaper than Islo?

In our scenarios, no: one coding agent on workdays comes to $228 a month on OpenAI and $101 on Islo, and one agent running all month comes to $1,051 versus $420. OpenAI also bills model tokens on top. bochka comes to $20 and $26 for the same two scenarios.

Which is faster?

We cannot say: neither OpenAI Containers nor Islo has a public CPU benchmark result, and OpenAI does not publish vCPU per container size. bochka did 26.69 runs per second on the Node.js build-and-test loop at 4 vCPU / 8 GB.

Can OpenAI Containers run Docker?

OpenAI's docs say hosted shell commands do not run with sudo, and no Docker support is documented. Islo runs Docker inside its microVMs, and bochka gives root, Docker, systemd and /dev/kvm on every sandbox.

Do they keep files between sessions?

An OpenAI container is deleted after 20 minutes without use, and its data cannot be recovered. Islo keeps the disk, and pause keeps memory and processes too. bochka saves the disk every minute and when the sandbox stops.

Can I limit what the code reaches on the internet?

Yes on both. Code Interpreter and Hosted Shell containers have no outbound access until an admin sets a domain allowlist, and Islo's gateway allows or denies by host, path and HTTP method and logs each request. bochka sandboxes have open internet with no provider-side allowlist.

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.

Try bochka

Install the CLI, then run bochka new.

Product preview