OpenAI Containers vs AWS AgentCore
OpenAI Containers are a tool inside OpenAI's model API: the model runs Python or shell commands in a box that, by default, expires after 20 minutes without use. AWS AgentCore is an agent platform inside your AWS account: it hosts your agent code in a microVM per session, next to managed Code Interpreter and Browser tools.
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 a Responses call makes one. The model runs Python or shell commands in it on a fixed Debian 12 image without sudo. By default the container expires after 20 minutes without use; the create call can set a different idle time. 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.
AWS AgentCore
Amazon Bedrock AgentCore is a set of AWS services to deploy and run AI agents: a Runtime that hosts your agent code, plus built-in Code Interpreter and Browser tools.
AgentCore is not a sandbox product first. It is a hyperscaler's agent platform: you push a container image or code, and each Runtime session gets its own Firecracker microVM. You reach a session through AgentCore APIs, an interactive shell API and authenticated HTTPS session endpoints. Around it sit managed tools (Code Interpreter, Browser) and services for memory, identity, gateways and evaluations. An Instances compute type runs longer sessions on EC2 in your own account. It is built for teams who already build on AWS and want agents to stay inside their account, IAM and VPC.
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. “357th cheapest of 365” means 356 of the other 364 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 | AWS AgentCore: 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 stays. Hosted OpenAI containers run without sudo and discard files when they expire. An AgentCore microVM session has 2 vCPU, stops after 8 hours and keeps only files under /mnt. boat gives a 4 to 16 vCPU VM with Docker and saves its disk every minute. | 357th of 365, with limits $17,371 a month | 132nd of 365, with limits $888 a month | 5th cheapest of 365 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An assistant that lives on one machine all month outlasts both. An OpenAI container expires after its idle time (20 minutes by default, settable with expires_after). AgentCore needs the Instances type on EC2 in your account, up to 14 days per session. On boat you turn auto-stop off on a paid plan (default 1 hour). | 344th of 361, with limits $13,140 a month | 109th of 361, with limits $1,300 a month | 6th cheapest of 361 $657 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Both are built for short code runs. AgentCore Code Interpreter bills only CPU actually used and peak memory. OpenAI Containers fit best when the model already runs in an OpenAI API call. | 198th of 336, with limits $6,000 a month | 11th cheapest of 336 $114 a month | 189th of 336, 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 OpenAI documents no inbound ports for hosted containers. AgentCore session endpoints reach a session's app ports, but each request needs an AWS signature or an OAuth token. boat gives each sandbox up to 50 HTTPS URLs, token-gated by default or public. | 349th of 349, with limits $3,780 a month | 205th of 349, with limits $506 a month | 3rd cheapest of 349 $209 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 195th of 349, with limits $30,000 a month | 129th of 349, with limits $8,075 a month | 39th of 349, with limits $2,000 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 355th of 359, with limits $5,760 a month | 46th cheapest of 359 $650 a month | 10th cheapest of 359 $200 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month CI jobs need Docker and open internet. Code Interpreter containers start with no internet until an admin allows domains, and OpenAI documents no Docker. AgentCore documents no Docker or /dev/kvm inside a session. boat gives root with both. | 347th of 374, with limits $1,728 a month | 51st cheapest of 374 $399 a month | 2nd cheapest of 374 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB A dev box needs SSH and a disk that keeps installed packages. OpenAI documents no SSH for hosted containers. AgentCore offers an interactive shell API but no SSH, and keeps only files under /mnt after a stop. boat has SSH and saves its disk every minute. | 365th of 371, with limits $1,051 a month | 242nd of 371, with limits $52 a month | 19th cheapest of 371 $26 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month OpenAI and AgentCore both drive a browser for computer use. boat streams a full Linux desktop at 60fps for agents that use other apps too. | 353rd of 367, with limits $5,280 a month | 146th of 367, with limits $202 a month | 7th cheapest of 367 $200 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month AgentCore Browser is a managed Chromium with live view and session replay. OpenAI's hosted browser is part of its Agents API beta. boat runs Chrome on a full desktop you control. | 335th of 356, with limits $1,440 a month | 4th cheapest of 356 $144 a month | 7th cheapest of 356 $200 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 334th of 343, with limits $1,755 a month | 32nd cheapest of 343 $287 a month | 6th cheapest of 343 $90 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 307th of 327, with limits $35,100 a month | 164th of 327, with limits $5,458 a month | 73rd of 327, with limits $1,800 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month | 310th of 351, with limits $870 a month | 108th of 351, with limits $294 a month | 3rd cheapest of 351 $200 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month | Not ranked: misses a requirement | Not ranked: misses a requirement | Not ranked: misses a requirement |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Not ranked: misses a requirement | Not ranked: misses a requirement | Not ranked: misses a requirement |
Key features
| Feature | OpenAI Containers | AWS AgentCore | boat |
|---|---|---|---|
| What it is | A sandbox tool the model uses inside OpenAI's API | An AWS agent platform: hosted agent sessions plus managed Code Interpreter and Browser tools | A cloud VM per agent that you create, keep, stop and reach from outside |
| Isolation | A fully sandboxed virtual machine; hypervisor not named | Firecracker microVM per Runtime session; Code Interpreter and Browser in a containerized environment | Full VM with its own kernel on AMD Ryzen hosts dedicated to boat; vCPUs are shared |
| Machine size | Picked by memory: 1, 4, 16 or 64 GB; vCPU not published (Agents API sandboxes: 1, 2 or 4 vCPU) | Up to 2 vCPU / 8 GB per microVM session, not adjustable; on Instances the EC2 type sets the size | 4 vCPU / 8 GB by default, up to 16 vCPU / 32 GB on the $100 plan and above |
| Session length | Expires after 20 minutes without use by default; expires_after can set a longer idle time | Up to 8 hours per microVM lifecycle, stopped after 15 idle minutes by default; up to 14 days on Instances | Auto-stop defaults to 1 hour; set up to 30 days, or off on a paid plan for no cap |
| Files after a stop | Data is discarded when the container expires; the Agents API keeps files in /workspace/outputs as downloadable artifacts | Mounts under /mnt only: a 1 GB per-session mount (preview), your own EFS or S3 Files in a VPC, or EBS volumes on Instances; the rest resets | The whole disk, saved every minute and at stop; memory is not saved; kept at least 14 days after a stop |
| Root, Docker and /dev/kvm | Hosted commands do not run with sudo; no Docker or /dev/kvm documented | Root inside the microVM; Docker and /dev/kvm not documented | Root, systemd, Docker and /dev/kvm on every machine |
| Image | Fixed Debian 12 image; you upload files or skills into each container | Your own container image: arm64 on microVMs, x86_64 or arm64 on Instances | boat's image or a saved snapshot; you cannot boot your own Docker image |
| Snapshots and fork | No snapshot or fork call documented | No snapshot or fork of a session; on Instances you can seed EBS volumes from your own EBS snapshots | Automatic disk snapshot every minute; fork a running sandbox from its last snapshot |
| Inbound access | No inbound ports documented for hosted containers; you reach them through the API | Authenticated HTTPS session endpoint for 1 to 5 ports (IAM or OAuth on every request, not on Instances); interactive shell API; no SSH or raw TCP | SSH, up to 50 HTTPS URLs token-gated by default or public, raw ports over public IPv6 |
| Outbound network | Code Interpreter and Hosted Shell: off until an admin sets an org domain allowlist; Agents API sandboxes open by default, can be limited to 1 to 100 hosts | Open internet, or VPC mode with security groups, PrivateLink and a NAT gateway for a fixed outbound IP | Open internet; no provider-side allowlist; the IP changes at every resume |
| Browser and desktop | Hosted browser for the Agents API computer_use tool | Managed Chromium with live view, session recording and replay, profiles and proxies; no desktop beyond the browser | Full Linux desktop and Chrome streamed at 60fps, with computer tools and an embeddable stream |
| Coding agents | Agents API (beta) drives a managed Codex harness from outside, with steering, events and webhooks | AWS's managed harness built on Strands Agents, or your own agent code in any framework | Seven harnesses installed, among them Claude Code and Codex, with a prompt, steer and events API |
| Keeping API keys out of the sandbox | domain_secrets: a sidecar adds the real header for approved domains, the model sees a placeholder | AgentCore Identity stores OAuth tokens and API keys and hands them to your agent code | Optional AI credentials proxy ($1 a day plus $1 per 1,000 requests) for supported providers; other secrets still enter |
| GPUs and regions | No GPUs; Code Interpreter processing in the US or Europe, Agents API in the US | NVIDIA G-series and Inferentia2 on Instances; 21 commercial AWS regions plus GovCloud | No GPUs; Germany, Finland and France only, no US or Asia region |
Every feature Battleships tracks (43)
| Feature | OpenAI Containers | AWS AgentCore | boat |
|---|---|---|---|
| Full VM (own kernel) | Yes | Yes | Yes |
| Docker inside | No documented way | Not published | Yes |
| Nested virtualization | No documented way | No documented way | Yes |
| Root | No documented way | Yes | Yes |
| Systemd | Not published | No documented way | Yes |
| Desktop GUI | Yes | No documented way | Yes |
| Browser | Yes | Yes | Yes |
| Computer-use API | Yes | Yes | Yes |
| SSH | No documented way | No documented way | Yes |
| Persistent disk | No documented way | Yes | Yes |
| Pause/resume | No documented way | No documented way | Yes |
| Automatic snapshots | No documented way | No documented way | Yes |
| Fork/clone | No documented way | No documented way | Yes |
| Live fork (no pause) | No documented way | No documented way | Yes |
| Memory fork | No documented way | No documented way | No documented way |
| Custom image (Docker or snapshot) | No documented way | Yes | Yes |
| ≥24 h sessions | Yes | No documented way | Yes |
| Idle auto-stop | Yes | Yes | No documented way |
| Wake on request | No documented way | Yes | No documented way |
| HTTPS preview URLs | No documented way | Yes | Yes |
| Raw TCP inbound | No documented way | No documented way | Yes |
| Public IPv4 | No documented way | No documented way | No documented way |
| Egress allowlist | Yes | Yes | No documented way |
| Secret proxy | Yes | No documented way | Yes |
| Secret proxy for any API | Yes | No documented way | Yes |
| Preinstalled agents | No documented way | No documented way | Yes |
| Agent harness API | Yes | No documented way | Yes |
| MCP server | No documented way | Yes | No documented way |
| GPU | No documented way | Yes | No documented way |
| Windows | No documented way | No documented way | No documented way |
| Arm64 | No documented way | Yes | No documented way |
| Self-hosting | Yes | No documented way | No documented way |
| BYOC | No documented way | Yes | No documented way |
| Open source | No documented way | No documented way | No documented way |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | Not published | With your own setup | Yes |
| Usage per sandbox (API) | No documented way | No documented way | Yes |
| Audit logs | Not published | Yes | No documented way |
| SSO | Yes | Yes | No documented way |
| SOC 2 | Yes | Yes | No documented way |
| HIPAA | Yes | Yes | No documented way |
| EU data residency | Yes | Yes | No documented way |
| Zero data retention | Not published | Not published | Yes |
Performance
Neither OpenAI Containers nor AWS AgentCore is in the public benchmarks we use, so we cannot say which is faster, and OpenAI does not publish the vCPU behind each container size. An AgentCore microVM session stops at 2 vCPU. boat did 28.06 V8 Web Tooling Benchmark runs/s at 4 vCPU / 8 GB, second of 15 in StarSling's run of 3 October 2026 (run 37073680030). Every boat sandbox in this run ran on an AMD Ryzen 9 9950X host. boat also places some sandboxes on AMD Ryzen 7 9700X hosts, which this run did not measure.
V8 Web Tooling Benchmark at a 4 vCPU / 8 GB target, runs per second, higher is better. Source: StarSling's run of 3 October 2026 (run 37073680030), median of 5 sandboxes. Providers ran on different hardware in different regions. StarSling gives a shared rank when the run cannot separate two providers. Every boat sandbox in this run ran on an AMD Ryzen 9 9950X host. boat also places some sandboxes on AMD Ryzen 7 9700X hosts, which this run did not measure. OpenAI Containers and AWS AgentCore are not in the public benchmarks we use, so we cannot say which is faster.
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. Start limits, per-start fees and plan minimums are in the price, boat's included.
| Scenario | OpenAI Containers, a month | AWS AgentCore, 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 | $26 · Pay-as-you-go (Instances) · with limits | $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 | $107 · Pay-as-you-go (Instances) · with limits | $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 | $1,286 · Pay-as-you-go (Instances) · with limits | $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 | $423 · Pay-as-you-go · 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 | $1,615 · Pay-as-you-go (Instances) | $8,800 · $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 on top. There is no plan fee. The scenarios use the 16 GB size ($1.44 an hour) for 8 GB needs, because OpenAI does not publish the vCPU per size.
AWS AgentCore: No monthly fee. Runtime microVMs, Code Interpreter and Browser bill per second for CPU actually used and peak memory. The 4 vCPU shape in these scenarios is the Instances type: the EC2 on-demand price plus a 12% management fee, with EBS storage (billed also while stopped) and data transfer on top. The scenario prices leave out EBS storage.
AWS AgentCore: Plans cost $20, $100, $500 or $2,000 a month. The whole plan price comes back as VM time for that month, and unused time does not carry over. A 4 vCPU / 8 GB sandbox costs $0.036 an hour, billed per second while it runs, busy or idle. A stopped sandbox has no compute charge, but saved snapshots above the included count and the AI credentials proxy still charge. The $20 plan includes 200 starts a day; extra starts cost $0.03 to $0.07.
Limits in those scenarios
OpenAI Containers fit none of the five scenarios as written: hosted containers discard files when they expire and document no Docker, Code Interpreter and Hosted Shell start with no internet, the vCPU per size is not published, and OpenAI does not confirm 300 or 1,000 containers at once. AWS AgentCore fits the 1,000 short jobs. In the others, files outside /mnt reset after a stop, the disk price is not in the bill, and the always-on agent hits the 14-day cap on Instances. boat fits four of the five. For the burst, the $2,000 plan bills starts above the 2,400 included a day ($8,800 a month), and starts above 270 a minute fail with 429.
| Scenario | OpenAI Containers | AWS AgentCore | boat |
|---|---|---|---|
| One coding agent, workdays |
|
| None for this scenario |
| One agent that never stops |
|
| None for this scenario |
| 50 agents in parallel |
|
| None for this scenario |
| Sandboxes for your users |
|
| None for this scenario |
| 1,000 short jobs at once |
|
|
|
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.
- Locked-down network by default. Code Interpreter and Hosted Shell reach nothing until an admin allows domains, and secrets for them stay hidden from the model.
- Nothing to deploy. OpenAI creates the container, runs the code and lets it expire, so there is no agent image, account setup or cleanup.
Use AWS AgentCore when
your agents must stay inside your AWS account and VPC, need a GPU or a US or Asia region, or are short code and browser tasks that fit the managed tools.
- It lives inside your AWS account. Teams built on AWS keep agents behind their own IAM, VPC, security groups and PrivateLink, in 21 commercial regions.
- GPUs and two-week sessions on Instances. Agents that need a GPU or a session of up to 14 days can run on EC2 in your account.
- Managed tools that bill for what they use. Code Interpreter and Browser charge for CPU actually used and peak memory, and Browser adds live view, session replay and saved profiles.
Where boat fits
Both of these suit short work: OpenAI's box expires after its idle time, and AgentCore's microVM has 2 vCPU and stops after 8 hours. If your agent does real work on a machine for hours or days, with installs, Docker, a dev server and files that must stay, boat gives it a whole VM. boat runs only in the EU, has no GPU and limits starts per plan.
A whole machine with root
OpenAI runs hosted commands without sudo, and neither OpenAI nor AgentCore documents Docker or /dev/kvm. The ways around it are a sandbox you host for OpenAI's Agents API, or AgentCore Instances on EC2 in your account. Every boat sandbox has root, systemd, Docker and /dev/kvm.
A disk that stays, with fork
An OpenAI container discards its files when it expires. AgentCore keeps only paths under /mnt, so packages installed elsewhere must be installed again after a stop. boat saves the whole disk every minute and can fork it. A boat resume is like a reboot, and a machine failure loses writes after the last good snapshot.
A link anyone can open
OpenAI documents no inbound ports for hosted containers. AgentCore session endpoints need an AWS signature or OAuth token on every request. boat gives each sandbox up to 50 HTTPS URLs, token-gated by default or public, plus SSH.
Use boat when your agent needs a real machine it keeps for hours or days: root, Docker, a disk saved every minute, open internet and URLs you can share, with seven coding agents installed. Compare it one to one: boat vs OpenAI Containers and boat vs AWS AgentCore.
boat's own limits
- Plans start at $20 a month. The plan price comes back as VM time for that month. Unused time does not carry over. Source
- Starts (create, fork, resume): the $20 plan includes 12 a minute, 60 an hour and 200 a day. The $2,000 plan includes 90 a minute, 600 an hour and 2,400 a day. Extra starts cost $0.03 to $0.07. A start fails with 429 above 3 times the included starts per minute, or 6 times per hour or per day. Source
- A stopped sandbox has no compute charge. boat keeps its disk for at least 14 days. A saved snapshot keeps it longer; extra saved snapshots and the AI credentials proxy are paid. Source
- Auto-stop defaults to 1 hour. You can turn it off on a paid plan. The trial allows 2 hours at most. Source
- boat saves the disk every minute and at each stop. Memory, running programs and /tmp are not saved: a resume is like a reboot. A machine failure loses writes after the last good snapshot. Source
- Machines run in Germany, Finland and France only. Snapshot copies can be stored outside the EU. Source
- Public IPv6 and outbound IPv4. No dedicated IPv4. The IP changes at each resume. Source
- Hosts are AMD Ryzen 7 9700X and Ryzen 9 9950X. vCPUs are shared. No GPU. You cannot boot your own Docker image. Source
Questions people ask
Is OpenAI Containers cheaper than AWS AgentCore?
In every scenario on this page, AgentCore costs less. One coding agent on workdays is $228 a month on OpenAI and $26 on AgentCore, and one agent running all month is $1,051 against $107. OpenAI also bills model tokens on top, and AgentCore's Instances prices leave out EBS storage. boat comes to $20 and $26 a month for those two.
Which is faster, OpenAI Containers or AWS AgentCore?
Neither is in the public benchmarks we use, so we cannot say. AgentCore microVM sessions stop at 2 vCPU, and OpenAI does not publish vCPU per container size. boat did 28.06 V8 Web Tooling runs/s at 4 vCPU / 8 GB in StarSling's run of 3 October 2026, on a Ryzen 9 9950X host.
Can OpenAI Containers or AWS AgentCore run Docker?
OpenAI's hosted shell runs commands without sudo, and Docker is not documented; the Agents API can drive a sandbox you host. AgentCore runs your container image as root in a microVM, but its docs do not say Docker or /dev/kvm work inside a session. boat gives root, systemd, Docker and /dev/kvm on every sandbox.
How long can a session last on each?
An OpenAI container expires after 20 minutes without use by default, and expires_after can set a longer idle time. An AgentCore microVM session lasts up to 8 hours and stops after 15 idle minutes by default, or up to 14 days on Instances. boat has no session cap once auto-stop is off on a paid plan (default 1 hour).
Do files survive when the session ends?
On OpenAI, the container's data is discarded when it expires. On AgentCore, only paths under /mnt survive: a 1 GB per-session mount (preview), your own EFS or S3 Files in a VPC, or EBS volumes on Instances. boat saves the whole disk every minute and keeps it at least 14 days after a stop.
Can I open a web app running inside either one?
Not on a hosted OpenAI container: OpenAI documents no ports or SSH for them. On AgentCore, session endpoints reach the app, but each request needs an AWS signature or an OAuth token. boat gives each sandbox up to 50 HTTPS URLs, token-gated by default or public, plus SSH.
Try boat
Install the CLI, then run boat new.
7-day free trial
More comparisons
- boat vs OpenAI Containers
- boat vs AWS AgentCore
- E2B vs OpenAI Containers
- Daytona vs OpenAI Containers
- Modal vs OpenAI Containers
- Vercel Sandbox vs OpenAI Containers
- Cloudflare Sandbox vs OpenAI Containers
- Blaxel vs OpenAI Containers
- Runloop vs OpenAI Containers
- Tensorlake vs OpenAI Containers
- Morph Cloud vs OpenAI Containers
- CodeSandbox SDK vs OpenAI Containers