Koyeb vs Docker Cloud Sandboxes

Both run each sandbox in its own microVM. Koyeb is a serverless cloud whose sandboxes can sleep, wake on a request and run on GPUs in many regions, while Docker Cloud Sandboxes takes the sbx coding-agent sandbox from your laptop to the cloud, for up to 24 hours at a time and 10 at once by default.

The companies and products

Koyeb

Koyeb is a serverless cloud for apps, inference and sandboxes, and in February 2026 it agreed to join Mistral AI.

A Koyeb sandbox is a Koyeb service: a microVM started from any Docker image and driven from Python or JavaScript SDKs that run commands, read files and expose a port. It can scale to zero when idle and wake on the next request, and it can run on CPU or GPU instance types. It is built for teams that run AI agents or untrusted code next to apps they already host on Koyeb.

Docker Cloud Sandboxes

Docker Cloud Sandboxes runs Docker's sbx coding-agent microVMs on Docker-managed compute instead of your laptop.

Each sandbox is a microVM with its own Linux kernel and its own Docker Engine, started from the same sbx CLI you use locally, with Claude Code, Codex and other agents ready to run. You can move a sandbox from your laptop to the cloud and back, publish HTTPS ports and SSH in. It launched on September 24, 2026, is billed pay-as-you-go on a personal Docker account, and is built for developers who want their local agent sandboxes to keep going off their machine.

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. “4th cheapest of 345” means 3 of the other 344 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 workloadKoyeb: price rankDocker Cloud Sandboxes: price rankboat: price rank
Software factory
100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month
A software factory needs agents that run for days with Docker and a disk that stays; Docker stops each sandbox at 24 hours, and Koyeb's local disk is lost on redeploy.
54th of 367, with limits
$366 a month
280th of 367, with limits
$4,928 a month
3rd cheapest of 367
$634 a month
Always-on AI employees
50 machines 24/7 · 2 vCPU / 4 GB
135th of 366, with limits
$923 a month
276th of 366, with limits
$5,110 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
Koyeb fits short code runs well, with small instances that scale to zero and wake on the next request, while Docker's 10 sandboxes at once leave it a compromise.
4th cheapest of 345
$66 a month
119th of 345, with limits
$233 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
Serving your own users needs many sandboxes at once: Koyeb's Scale plan allows 1,000 services, Docker allows 10 by default, and boat fits it best of the three.
6th cheapest of 361
$495 a month
136th of 361, with limits
$1,628 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
237th of 360, with limits
$1,631 a month
114th of 360, with limits
$7,000 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
Evals run many sandboxes side by side, which Koyeb and boat handle and Docker's default of 10 at once does not.
19th cheapest of 364
$429 a month
167th of 364, with limits
$1,400 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 that builds Android apps or VM images needs /dev/kvm, which boat exposes and neither Koyeb nor Docker documents.
142nd of 377, with limits
$88 a month
172nd of 377, with limits
$672 a month
4th cheapest of 377
$86 a month
24/7 dev boxes
1 machine 24/7 · 4 vCPU / 8 GB
A dev box that stays on all month cannot live in one Docker session, and Koyeb's priced eco instance has a 20 GiB disk; boat keeps a full disk with no session cap.
217th of 368, with limits
$37 a month
127th of 368, with limits
$204 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
Computer-use agents need a screen, and neither Koyeb nor Docker has a desktop in the sandbox; boat streams a 60fps Linux desktop with Chromium.
81st of 369, with limits
$123 a month
268th of 369, with limits
$1,400 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
172nd of 367, with limits
$123 a month
277th of 367, with limits
$700 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
172nd of 346, with limits
$89 a month
262nd of 346, with limits
$350 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
65th of 338, with limits
$1,683 a month
190th of 338, with limits
$7,000 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
233rd of 360, with limits
$299 a month
248th of 360, with limits
$350 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
Only Koyeb runs GPU instances, from RTX 4000 cards up to H100 and H200; Docker and boat are CPU only.
3rd cheapest of 98
$628 a month
Can't run itCan't run it
macOS agents
5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month
Can't run itCan't run itCan't run it

Key features

FeatureKoyebDocker Cloud Sandboxesboat
Isolation and DockerMicroVM per instance, started from any Docker imageMicroVM with its own Linux kernel and its own Docker Engine; agent runs as a non-root user with sudoFull VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts, with root, systemd and Docker
Machines and GPUsUp to 40 vCPU / 128 GB, plus GPUs from RTX 4000 SFF Ada up to H100 and H2001 vCPU / 2 GiB up to 16 vCPU / 32 GiB, CPU only, amd64 or arm64 by accountUp to 16 vCPU / 32 GB on the $100 plan and above; no GPUs
Session lengthNo forced cap; auto-delete timers are optional1 hour by default, at most 24 hours from creationNo cap on paid plans; you set the time to live
Sandboxes at once100 services on Pro and 1,000 on Scale, each sandbox being a service10 by default, 50 stored; no published way to raise it100 on the $20 plan, up to 2,000 on higher plans, self-serve
Disk that survivesLocal disk is lost on redeploy; volumes (preview, 1 to 10 GB, two regions) persistStopped state is kept, but snapshots unused for 7 days are deleted; disk size not publishedDisk kept across stops and saved every minute, with snapshots
Sleep and resumeScales to zero after at least 5 idle minutes and wakes on the next request; Light Sleep wakes in about 200 msStop keeps memory and files; resume picks up with processes still runningStop and start by API; a stopped sandbox costs nothing and keeps its disk
Snapshots and forkSDK snapshots of files, or with running processes, then new sandboxes from them (experimental)Snapshot a running sandbox with disk or disk and memory, then restore it as a new sandboxFork a running sandbox with its disk
Nested virtualizationNot documentedNot documentedYes, /dev/kvm for Android emulators and VMs
Inbound accessOne exposed HTTPS port at a time, plus an optional TCP proxy to one port; no SSHPublic HTTPS URL per published port, not a TCP tunnel; SSH with Docker-issued certificatesHTTPS URLs for up to 50 ports, token-protected by default; raw TCP; dedicated IPv4 or IPv6
Outbound rulesBlock all outbound traffic, or allow only listed IPs and CIDRsAllow and deny rules by host, with allow-all, balanced and deny-all presetsOpen internet, no provider-side allowlist
Desktop and browserNo desktop or browser in the sandbox SDKTerminal sessions only60fps Linux desktop with Chromium and computer use
Coding agentsTutorials that run agents through the command API; an MCP server lets agents manage KoyebClaude Code, Codex, Copilot, OpenCode and more, run and attached to from the sbx CLISeven agents preinstalled, with a prompt, steer and events API to drive them from outside
Keeping API keys out of the sandboxSecrets are set as environment variables inside the serviceCloud proxy sets the auth header for built-in AI services and custom hosts; the sandbox sees a placeholderCredentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out
Where it runsFrankfurt, Paris, Washington D.C., AWS US East, Singapore, Tokyo; on-premise for EnterpriseRegions not published; the same microVM also runs on your laptopEU only (Germany, Finland, France)
Every feature Battleships tracks (43)
FeatureKoyebDocker Cloud Sandboxesboat
Full VM (own kernel)YesYesYes
Docker insideYesYesYes
Nested virtualizationNoNoYes
RootNot publishedYesYes
SystemdNot publishedNot publishedYes
Desktop GUINoNoYes
BrowserNoNoYes
Computer-use APINoNoYes
SSHNoYesYes
Persistent diskYesYesYes
Pause/resumeYesYesYes
Automatic snapshotsYesYesYes
Fork/cloneYesYesYes
Live fork (no pause)Not publishedYesYes
Memory forkYesYesNo
Custom image (Docker or snapshot)YesYesYes
≥24 h sessionsYesYesYes
Idle auto-stopYesNoNo
Wake on requestYesNot publishedNo
HTTPS preview URLsYesYesYes
Raw TCP inboundYesNot publishedYes
Public IPv4NoNoNo
Egress allowlistYesYesNo
Secret proxyNoYesYes
Secret proxy for any APINoYesYes
Preinstalled agentsYesYesYes
Agent harness APINoNoYes
MCP serverYesYesNo
GPUYesNoNo
WindowsNoNoNo
Arm64NoYesNo
Self-hostingYesNoNo
BYOCYesNoNo
Open sourceNoNoNo
Scoped API keysNot publishedNot publishedYes
Spend limitsNoNot publishedYes
Usage per sandbox (API)NoNoYes
Audit logsYesNot publishedNo
SSOYesYesNo
SOC 2YesNoNo
HIPAANoNoNo
EU data residencyYesNoYes
Zero data retentionNot publishedNot publishedYes

Performance

KoyebNot measured
Docker Cloud SandboxesNot measured
boat26.69 runs/s

There is no comparable measurement for Koyeb or for Docker Cloud Sandboxes: neither is in the hpc-sandbox-benchmarks runs, and Docker does not say what hardware it uses. Koyeb publishes a CPU share of at least 0.25 dedicated vCPU per GB of memory, bursting to all cores. boat did 26.69 runs per second on that Node.js build-and-test benchmark at 4 vCPU / 8 GB.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Koyeb and Docker Cloud Sandboxes 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.

ScenarioKoyeb, a monthDocker Cloud Sandboxes, a monthboat, 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
$29 · Pro$49 · Docker Agentic Platform (pay-as-you-go)$20 · $20/mo
One agent that never stops
4 vCPU / 8 GB, 50 GB disk, running 24/7 all month
$57 · Pro$204 · Docker Agentic Platform (pay-as-you-go) · 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
$475 · Pro$2,464 · Docker Agentic Platform (pay-as-you-go) · 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
$329 · Scale$700 · Docker Agentic Platform (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
$485 · Scale · with limits$1,400 · Docker Agentic Platform (pay-as-you-go) · with limits$2,000 · $2000/mo · with limits

Koyeb: Koyeb bills each instance per second at a fixed price per instance type, and Deep Sleep costs nothing. Pro costs $29 a month with $10 of usage included and Scale $299 a month with $100 included; the rest of the fee is not credit.

Docker Cloud Sandboxes: Docker bills per second for the size you pick, from $0.07 an hour for 1 vCPU / 2 GiB to $1.12 an hour for 16 vCPU / 32 GiB, with no monthly fee. A stopped sandbox costs nothing, and volumes and egress are free.

Docker Cloud Sandboxes: 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.

Limits in those scenarios

Docker's limits bite first: its 24-hour session cap rules out one agent that never stops, and its default of 10 sandboxes at once rules out 50 agents in parallel, sandboxes for your users and 1,000 short jobs at once. Koyeb fits all but the burst, though its priced eco-xlarge shape has a 20 GiB disk with no published price for more. At 1,000 short jobs at once nothing fits as priced: Koyeb's eco CPUs guarantee half of each vCPU and throttle above that, and boat queues starts above 90 a minute.

ScenarioKoyebDocker Cloud Sandboxesboat
One coding agent, workdays
  • disk beyond 20 GiB: price unknown
None that biteNone that bite
One agent that never stops
  • disk beyond 20 GiB: price unknown
  • Docker Agentic Platform (pay-as-you-go): sessions capped at 24 h
None that bite
50 agents in parallelNone that bite
  • Docker Agentic Platform (pay-as-you-go): max 10 concurrent
None that bite
Sandboxes for your usersNone that bite
  • Docker Agentic Platform (pay-as-you-go): max 10 concurrent
None that bite
1,000 short jobs at once
  • burstable CPU: 50% of each vCPU guaranteed, throttled above that
  • Docker Agentic Platform (pay-as-you-go): max 10 concurrent
  • 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 Koyeb when

your sandboxes need GPUs, sleep most of the time and must wake on a request, run in the US or Asia, or sit next to apps you already host on Koyeb.

  • GPUs in the same platform. Useful when an agent or eval needs a GPU: Koyeb runs instances from RTX 4000 cards up to H100 and H200.
  • Sleep when idle, wake on request. Good for sandboxes that sit idle between requests: Koyeb scales them to zero and wakes them on the next call.
  • No session cap, many regions. Koyeb sandboxes can run until you delete them, in the EU, the US or Asia.

Use Docker Cloud Sandboxes when

you already run agents in sbx on your laptop and want a few of them to keep going in the cloud for up to a day, with memory kept across stops.

  • The same sandbox on your laptop. If you already run sbx locally, the cloud is the same microVM and CLI, and you can move a sandbox either way.
  • Stop and fork with memory. Useful for agents that pause between steps: a stopped sandbox comes back with its memory and running processes.
  • Keys and outbound rules at the proxy. A cloud proxy adds API keys outside the VM and applies allow and deny rules by host.

Where boat fits

If you are choosing between Koyeb and Docker Cloud Sandboxes to run coding agents, look at boat when those agents work for hours or days on one machine. boat gives each one a full VM that keeps its disk, has no session cap on paid plans, and comes with a desktop, hosting and an API to drive the agents inside from outside.

A disk that stays, with no time limit

On Koyeb the local disk is lost on redeploy, so lasting files need a volume in preview at 1 to 10 GB. Docker ends every sandbox 24 hours after creation and deletes snapshots unused for 7 days.

A whole machine, screen included

Neither Koyeb nor Docker documents /dev/kvm or offers a desktop, so Android emulators, VMs and computer-use agents need another service. boat has /dev/kvm and a 60fps Linux desktop with Chromium in every sandbox.

Agents and ports driven from outside

On Koyeb you build the agent harness on exec and expose one port at a time; on Docker you attach to a terminal and parse its text. boat has a prompt, steer and events API and HTTPS on 50 ports.

Use boat when your agents do real work for hours or days on a machine that keeps its disk: builds, Docker, emulators, a desktop browser and preview servers, driven from your own app through an API. Compare it one to one: boat vs Koyeb and boat vs Docker Cloud Sandboxes.

Questions people ask

Is Koyeb cheaper than Docker Cloud Sandboxes?

For one coding agent on workdays, yes: $29 against $49 a month for a 4 vCPU / 8 GiB machine. Koyeb also fits one agent that never stops at $57 a month and 50 agents in parallel at $475 a month, which Docker does not fit because of its 24-hour cap and 10 sandboxes at once.

Which is faster, Koyeb or Docker Cloud Sandboxes?

Neither has a comparable CPU measurement: both are missing from the hpc-sandbox-benchmarks runs. Koyeb guarantees at least 0.25 dedicated vCPU per GB of memory, and Docker does not say what hardware it uses. boat did 26.69 runs per second on that benchmark.

Can a Docker cloud sandbox run longer than 24 hours?

Not in one session. It expires after one hour by default, and extensions cannot go past 24 hours from creation. A Koyeb sandbox has no forced cap, and neither does a boat sandbox on paid plans.

Can I run Docker inside Koyeb or Docker Cloud Sandboxes?

A Docker cloud sandbox has its own Docker Engine inside the microVM. A Koyeb sandbox starts from any Docker image. boat runs Docker inside the sandbox with root and systemd, and also exposes /dev/kvm for Android emulators and VMs.

Which one has GPUs?

Only Koyeb. It runs instances from an RTX 4000 SFF Ada up to H100 and H200, billed per second. Docker Cloud Sandboxes and boat are CPU only.

How many sandboxes can I run at once?

Docker allows 10 at once and 50 stored per account by default. Koyeb counts each sandbox as a service, with 100 on Pro and 1,000 on Scale. boat runs 100 at once on the $20 plan and up to 2,000 on higher plans.

Try boat

Install the CLI, then run boat new.

7-day free trial

More comparisons