Koyeb vs Agent 37
Koyeb is a serverless cloud whose sandboxes are microVMs that scale to zero, run on GPUs and live in many regions. Agent 37 hosts one always-on agent per user in lighter gVisor sandboxes in the US, which sleep with their memory kept and cost less per hour.
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.
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.
- Where it runs
- Dedicated servers in the United States
- Instance types
- Default (shared), Performance (dedicated cores), VM
- Compliance
- SOC 2 Type I, Type II in progress
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 workload | Koyeb: 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 software factory needs Docker and a disk that lasts; Koyeb loses local disk on redeploy unless you add a volume, Agent 37 needs its VM type for Docker, and boat has both by default. | 54th of 367, with limits $366 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 This is what Agent 37 is built for: one assistant per user that sleeps when idle, keeps its memory and wakes on the next message; on Koyeb a deep sleep throws away the VM. | 135th of 366, with limits $923 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 Both fit short code runs well: Agent 37's small default sandboxes are the lowest bill, and Koyeb scales to zero after 5 minutes and wakes on the next request. | 4th cheapest of 345 $66 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 Koyeb fits this one with sandboxes that wake on request in many regions, while Agent 37 tops out at 200 instances on self-serve, counting sleeping ones; boat plans go up to 2,000 at once. | 6th cheapest of 361 $495 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 | 237th of 360, with limits $1,631 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 | 19th cheapest of 364 $429 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 | 142nd of 377, with limits $88 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 is built to keep a machine around the clock cheaply; Koyeb's eco shape caps disk at 20 GiB, and boat keeps a full VM with systemd and raw TCP. | 217th of 368, with limits $37 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 Agent 37 adds a noVNC desktop through a template build, Koyeb sandboxes have no desktop, and boat streams a 60fps desktop with Chromium in every sandbox. | 81st of 369, with limits $123 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 On Koyeb you install and run a headless browser yourself; Agent 37 and boat both ship a browser on a desktop, and boat's runs in every sandbox. | 172nd of 367, with limits $123 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 | 172nd of 346, with limits $89 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 | 65th of 338, with limits $1,683 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 | 233rd of 360, with limits $299 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 Koyeb is the only one of the three with GPUs, from RTX 4000 cards up to H100 and H200. | 3rd cheapest of 98 $628 a month | 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 | Koyeb | Agent 37 | boat |
|---|---|---|---|
| Isolation | MicroVM per instance | 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 |
| Docker | Sandboxes start from any Docker image | Docker inside 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 | Not documented | /dev/kvm on the VM type, with a ready Android template | /dev/kvm in every sandbox, Android emulator works |
| CPU | At least 0.25 dedicated vCPU per GB of memory, bursting to all cores; eco shapes guarantee half of each vCPU; not benchmarked | Shared by default; Performance type on dedicated cores at 4x the compute rate; not benchmarked | vCPUs on dedicated AMD Ryzen 9 9950X hosts; 26.69 runs/s on the CPU benchmark |
| Disk that survives | Local disk is lost on redeploy; volumes (preview, 1 to 10 GB, two regions) persist | Disk kept while the instance exists, billed even while stopped or asleep | Disk kept across stops and saved every minute, with snapshots |
| Sleep and wake | Scales to zero after 5 idle minutes or more and wakes on the next request; Light Sleep wakes in about 200 ms | Opt-in auto-sleep after a quiet period (default 15 minutes), memory checkpointed, wakes on the next request | No idle timer; stop and start by API, a stopped sandbox costs nothing and keeps its disk |
| Snapshots and fork | SDK snapshots of the filesystem or running processes, then new sandboxes from a snapshot (experimental) | 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 |
| Inbound access | One exposed HTTPS port at a time, plus an optional TCP proxy to one port; no dedicated 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 by default; raw TCP; dedicated IPv4 or IPv6 |
| Outbound rules | Block all outbound traffic, or allow only listed IPs and CIDRs | Open internet, no allowlist | Open internet, no provider-side allowlist |
| Desktop and browser | No desktop or browser in the sandbox SDK | noVNC desktop through a template build or the Ubuntu VM template | 60fps Linux desktop with Chromium and computer use in every sandbox |
| Coding agents | Tutorials that run agents through the command API; an MCP server lets agents manage Koyeb | Templates for Hermes, OpenClaw, Claude Code, Codex, Grok, OpenCode and Pi, with a chat API per instance | Seven agents preinstalled, including Claude Code and Codex, with a prompt, steer and events API |
| Keeping API keys out of the sandbox | Secrets are set as environment variables inside the service | 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 |
| GPUs | RTX 4000 SFF Ada up to H100, H200 and multi-GPU | No | No |
| Regions | EU, US and Asia | United States | EU only (Germany, Finland, France), EU data residency |
Every feature Battleships tracks (43)
| Feature | Koyeb | Agent 37 | boat |
|---|---|---|---|
| Full VM (own kernel) | Yes | No | Yes |
| Docker inside | Yes | No | Yes |
| Nested virtualization | No | No | Yes |
| Root | Not published | Yes | Yes |
| Systemd | Not published | Not published | Yes |
| Desktop GUI | No | Yes | Yes |
| Browser | No | Yes | Yes |
| Computer-use API | No | Yes | Yes |
| SSH | No | Yes | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | Yes | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Not published | Yes | Yes |
| Memory fork | Yes | No | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | Yes | No |
| Wake on request | Yes | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | Yes | Not published | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | Yes | No | No |
| Secret proxy | No | Yes | Yes |
| Secret proxy for any API | No | No | Yes |
| Preinstalled agents | Yes | Yes | Yes |
| Agent harness API | No | Yes | Yes |
| MCP server | Yes | No | No |
| GPU | Yes | No | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | Yes | No | No |
| BYOC | Yes | No | No |
| Open source | No | No | No |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | No | 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
There is no comparable measurement for Koyeb or Agent 37: neither was part of the hpc-sandbox-benchmarks run. boat did 26.69 runs per second on that Node.js build-and-test loop at 4 vCPU / 8 GB. Koyeb guarantees at least 0.25 dedicated vCPU per GB of memory, and Agent 37 shares cores by default but sells a Performance type on dedicated cores.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Koyeb 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 | Koyeb, 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 | $29 · Pro | $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 | $57 · Pro | $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 | $475 · Pro | $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 | $329 · Scale | $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 | $485 · Scale · with limits | $73 · Wallet ($250 lifetime top-ups) · 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.
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, plus $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: 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
Koyeb's priced shape is eco-xlarge, which has a 20 GiB disk with no published price for more, and eco instances guarantee only half of each vCPU, which throttles the 1,000-jobs burst. Agent 37 caps instances at 200 on its top self-serve tier, counting sleeping and stopped ones, so 300 user sandboxes at once and the burst do not fit; its 4 vCPU / 8 GB shape also caps disk at 20 GB and its egress price is not published. The 50-agent fleet needs Docker, so Agent 37 is priced there on its VM type at 2.5x compute.
| Scenario | Koyeb | Agent 37 | boat |
|---|---|---|---|
| One coding agent, workdays |
|
| None that bite |
| One agent that never stops |
|
| None that bite |
| 50 agents in parallel | None that bite |
| None that bite |
| Sandboxes for your users | None that bite |
| None that bite |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use Koyeb when
your sandboxes need GPUs, sleep most of the time and must wake on a request, or have to run in the US or Asia, maybe next to apps you already host there.
- 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.
- Many regions, fast wake. Good for sandboxes that must sit close to users in the EU, US or Asia and wake on a request in about 200 ms.
- Outbound rules. Koyeb can cut a sandbox off from the internet or allow only listed IPs, which helps when running untrusted code.
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 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.
- Agent templates with a chat API. Agent 37 starts Hermes, OpenClaw, Claude Code and other agents from templates, each with its own chat API.
- A low running bill. A prepaid wallet with no plan fee, metered per minute, gives the lowest bill here for one agent on workdays or always on.
Where boat fits
Koyeb is a serverless cloud that added sandboxes, and Agent 37 hosts mostly idle assistants. If your agents do hours of real work, building, testing and running Docker, emulators and preview servers, boat gives each one a whole machine that keeps its disk, with agents, a desktop and hosting built in.
A full VM by default
Every boat sandbox has root, systemd, Docker and /dev/kvm. Koyeb does not document /dev/kvm, and Agent 37's default gVisor type cannot run Docker; its VM type can, at 2.5x the compute rate and with no init system.
A disk that stays, and forks of a running machine
boat saves the disk every minute and can fork a sandbox while it runs. On Koyeb local disk is lost on redeploy unless you add a preview volume, and Agent 37 has no fork, only backups restored in place.
Keys outside, ports and a desktop built in
boat keeps real keys in a proxy for any HTTPS API and gives each sandbox 50 HTTPS ports and a 60fps desktop. Koyeb and Agent 37 put your own keys in env vars the agent can read, and Koyeb exposes one port at a time.
Use boat when your agents do real work for hours on a machine that keeps its disk: builds, Docker, emulators, a desktop browser and preview servers, with forks of running machines and keys kept outside for any API. Compare it one to one: boat vs Koyeb and boat vs Agent 37.
Questions people ask
Is Koyeb cheaper than Agent 37?
No, Agent 37's bill is lower in every scenario on this page that both fit: for example $12 against $57 a month for one agent that never stops, and $329 against $475 for 50 agents in parallel. Agent 37 does not fit 300 user sandboxes at once, because its self-serve tiers stop at 200 instances.
Which is faster, Koyeb or Agent 37?
Neither has a result on the hpc-sandbox-benchmarks run, so there is no fair speed comparison. boat did 26.69 runs per second there at 4 vCPU / 8 GB. Koyeb guarantees a share of each vCPU, and Agent 37 sells dedicated cores on its Performance type.
Can Koyeb and Agent 37 run Docker?
Koyeb starts each sandbox from any Docker image. Agent 37's default gVisor type cannot run a Docker daemon, but its VM type comes with Docker and Compose at 2.5x the compute rate. boat runs Docker in every sandbox.
Is an Agent 37 sandbox a VM like Koyeb's?
Not by default. Koyeb runs each instance in a microVM, while Agent 37's default is a container enforced by gVisor; you can ask for its VM type, a full Linux machine with its own kernel and KVM. Every boat sandbox is a full VM.
Which one has GPUs?
Only Koyeb, with instances from an RTX 4000 SFF Ada up to H100 and H200, billed per second. Agent 37 and boat are CPU only.
Where do Koyeb, Agent 37 and boat run?
Koyeb runs in Frankfurt, Paris, Washington D.C., Singapore, Tokyo and more. 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