Fly.io Machines vs Agent 37

Fly.io Machines are microVMs for hosting your own apps in 17 regions, and you build any agent setup yourself. Agent 37 is made for agents: one sandbox per user with agents ready, which sleeps when idle and keeps its memory.

The companies and products

Fly.io Machines

Fly.io is an app hosting platform, and Machines are the microVMs underneath it that you can also start and stop through a REST API.

A Fly Machine boots your Docker image as a Firecracker microVM on Fly's own servers, in one of 17 regions. You pick a shared or performance CPU preset, attach at most one Volume for data that must last, and Fly Proxy can stop, suspend and wake Machines as traffic comes and goes. It is built for teams that host web apps and services worldwide; it is not a sandbox product, so this page compares the parts that overlap with agent work.

Kind of product: App hosting platform; Machines are its VM building block with a REST API
Isolation: Firecracker microVMs with hardware virtualization, on Fly's own servers
Regions: 17 regions in North America, Europe, Asia, Oceania, South America and Africa
CPU classes: Shared vCPUs guarantee 6.25% of each core with a burst balance; performance vCPUs get the full core
Storage: Root file system is ephemeral; one Volume of up to 500GB per Machine for lasting data
Free trial: 2 Machine hours or 7 days, up to 10 Machines of 2 vCPU / 4GB that stop after 5 minutes
Compliance: Lists SOC 2 Type II; a HIPAA package with a pre-signed BAA is sold as an add-on

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. “92nd cheapest of 367” means 91 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	Fly.io Machines: 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 Agent 37 ranks 2nd and boat 3rd here; on Fly each agent needs its own Volume because the root disk resets, and there is no /dev/kvm.	92nd of 367, with limits $836 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 ranks 1st: an always-on assistant per user that sleeps when idle, keeps its memory and wakes on the next message is what it is built for.	125th of 366, with limits $816 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	75th of 345, with limits $94 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 boat ranks 1st; Agent 37 holds at most 200 instances on its top self-serve tier, counting sleeping ones, and Fly's API creates about one Machine per second per app.	69th of 361, with limits $600 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	48th of 360, with limits $2,838 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	117th of 364, with limits $457 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 keeps the CPU busy, which drains a shared Fly vCPU's burst balance; Agent 37 and boat both fit fully here.	152nd of 377, with limits $288 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	118th of 368, with limits $53 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, boat streams a 60fps desktop out of the box, and on Fly you build and stream a desktop yourself.	95th of 369, with limits $173 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 boat has Chromium and browser control in every machine; on Fly you package the browser into your own image.	160th of 367, with limits $90 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	161st of 346, with limits $67 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 Fly can spread scrapers across 17 regions and pay only for the seconds they run, while Agent 37 runs only in the US.	44th of 338, with limits $1,232 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 attempt to try many paths at once; neither Fly nor Agent 37 copies a running machine's disk into a new one.	118th of 360, with limits $63 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	Fly.io Machines	Agent 37	boat
Kind of product	App hosting platform; you boot your own image as a microVM	Agent hosting: one isolated sandbox per user behind one API	Agent sandbox: a full VM per agent, ready to use
Isolation	Firecracker microVM with its own kernel	gVisor by default; full VM as an opt-in instance type	Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts
Docker inside	You install it yourself in the microVM	Only on the VM type, at 2.5x the default compute rate	Ready in every sandbox
systemd and init	Fly's own init, no systemd	Your entrypoint is PID 1 with no init system, even on the VM type	systemd runs as in any Linux server
Nested virtualization	/dev/kvm not offered	/dev/kvm on the VM type, with an Android template	/dev/kvm in every sandbox
Files after a stop	Root file system is ephemeral; one Volume per Machine, billed for its full size all month	Disk kept, billed per GB even while stopped or asleep	Whole disk kept, snapshots every minute and on stop; a stopped sandbox costs nothing
Sleep and wake	Fly Proxy wakes Machines on request; memory kept on suspend only up to 2GB	Opt-in auto-sleep after a quiet period, memory checkpointed, wakes on the next request	No idle timer; resume reboots from disk
Fork	Clone copies a Machine's config; a cloned Volume starts empty	No fork; daily backups restored in place only	Fork a running sandbox, disk included
Desktop and computer use	None built in; you add it to your image	noVNC desktop through a template build	60fps desktop and Chromium, computer use, embeddable stream
Coding agents	None built in	Hermes, OpenClaw, Claude Code, Codex, Grok, OpenCode, Pi, with a chat API	Claude Code, Codex, OpenCode and more, with a prompt, steer and events API
Keys kept out of the sandbox	Secrets as env vars; open-source tokenizer proxy you run yourself	Its managed models, search and Composio via a token; your own keys as env vars	Proxy for model providers and any HTTPS API you register
Networking	HTTPS and TCP through Fly Proxy, Anycast IPs, custom domains, private network	HTTPS URL per port, custom domains; no public IP, SSH through a WebSocket tunnel	HTTPS URLs for up to 50 ports, raw TCP, dedicated IPv6 or IPv4
CPU	Shared vCPUs guarantee 6.25% of a core plus bursts; performance vCPUs get the full core	Shared by default; Performance type on dedicated cores at 4x the compute rate	One CPU class, no baseline quota or burst balance
Regions	17 regions on six continents	United States	EU only: Germany, Finland, France

Every feature Battleships tracks (43)

Feature	Fly.io Machines	Agent 37	boat
Full VM (own kernel)	Yes	No	Yes
Docker inside	Yes	No	Yes
Nested virtualization	No	No	Yes
Root	Yes	Yes	Yes
Systemd	No	Not published	Yes
Desktop GUI	No	Yes	Yes
Browser	No	Yes	Yes
Computer-use API	No	Yes	Yes
SSH	Yes	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)	Yes	Yes	Yes
Memory fork	No	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	Yes	No	No
Egress allowlist	Yes	No	No
Secret proxy	No	Yes	Yes
Secret proxy for any API	No	No	Yes
Preinstalled agents	No	Yes	Yes
Agent harness API	No	Yes	Yes
MCP server	Yes	No	No
GPU	No	No	No
Windows	No	No	No
Arm64	No	No	No
Self-hosting	No	No	No
BYOC	No	No	No
Open source	Yes	No	No
Scoped API keys	Not published	Not published	Yes
Spend limits	No	Yes	Yes
Usage per sandbox (API)	No	Yes	Yes
Audit logs	Not published	Not published	No
SSO	No	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

Fly.io MachinesNot measured

Agent 37Not measured

boat26.69 runs/s

There is no comparable measurement for either: neither Fly.io Machines nor Agent 37 was part of the public StarSling run where boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB. Both sell shared CPUs by default, Fly with a 6.25% baseline per vCPU, and both sell dedicated cores at a higher rate.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Fly.io Machines 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	Fly.io Machines, 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	$15 · 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	$36 · 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	$487 · 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	$110 · 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	$232 · Pay As You Go · with limits	$73 · Wallet ($250 lifetime top-ups) · with limits	$2,000 · $2000/mo · with limits

Fly.io Machines: Fly bills each Machine per second while it runs, with no plan fee; Volumes are billed for their full size all month. The table uses Fly's lowest rate: shared vCPUs bought in 1-year reservation blocks.

Agent 37: A prepaid wallet with no plan fee, metered per minute while running, plus disk per GB even while stopped or asleep. Performance compute costs 4x and VM compute 2.5x the default rate.

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

Fly is a compromise in all five scenarios: its lowest figures need a 1-year reservation and use shared vCPUs that guarantee only 6.25% of each core, so long builds get throttled. Agent 37 caps instances at 200 on its top self-serve tier, counting sleeping and stopped ones, so it does not fit 300 user sandboxes or 1,000 jobs at once, and it does not publish its egress price. The 50-agent fleet needs Docker, so Agent 37 is priced on its VM type there. boat queues the 1,000-job burst because of its limits on starts per minute and per day.

Scenario	Fly.io Machines	Agent 37	boat
One coding agent, workdays	different kind of product: app platform 1-year commitment burstable CPU: 6.25% of each vCPU guaranteed, throttled above that	egress price not published	None that bite
One agent that never stops	different kind of product: app platform 1-year commitment burstable CPU: 6.25% of each vCPU guaranteed, throttled above that	egress price not published	None that bite
50 agents in parallel	different kind of product: app platform 1-year commitment burstable CPU: 6.25% of each vCPU guaranteed, throttled above that	egress price not published	None that bite
Sandboxes for your users	different kind of product: app platform 1-year commitment burstable CPU: 6.25% of each vCPU guaranteed, throttled above that	max 200 concurrent egress price not published	None that bite
1,000 short jobs at once	different kind of product: app platform 1-year commitment burstable CPU: 6.25% of each vCPU guaranteed, throttled above that	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 Fly.io Machines when

you host apps or light services that must run near users outside Europe and wake on traffic, and you are happy to build the agent setup yourself.

17 regions worldwide. Useful if your users are spread out: Fly runs Machines in 17 regions, from the US to Sydney and Johannesburg.
A full app platform around each Machine. Useful if you host whole products: Fly adds network policies, a private network, Anycast IPs and custom domains.
Per-second billing, no plan fee. Useful for short bursts: Fly bills each Machine per second only while it runs, with no plan fee.

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 when idle 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.
Agents ready to run. Useful if you want agents now: templates for Hermes, OpenClaw, Claude Code, Codex and more, each with a chat API.
A VM type with /dev/kvm. Useful for Android emulators: Agent 37's VM type has Docker, KVM and a ready Android template, which Fly does not offer.

Where boat fits

Fly gives you a fast microVM and leaves the agent setup to you, and Agent 37 gives you agents in light sandboxes that sleep. If your agents do real work for hours, with installs, builds, Docker and emulators, boat gives each one a full VM with all of that ready, a disk that is always kept, and forks of running machines.

Docker, systemd and /dev/kvm by default

Fly has no systemd or /dev/kvm, and you install Docker yourself. Agent 37's default type cannot run Docker; its VM type can, but at 2.5x the compute rate and with no init system.

A disk that is kept, and forks of running machines

A Fly Machine loses anything outside its one Volume on restart, and a clone gets an empty Volume. Agent 37 has no fork, and its backups restore only onto the same instance.

Keys stay outside for any API

On Fly, keys go in as env vars unless you run your own tokenizer app. Agent 37 keeps only its managed models, search and Composio outside; your own keys sit inside the machine the agent controls.

Use boat when your agents do real work for hours: installs, builds, Docker and emulators on a full VM whose disk is always kept, with forks, a desktop and a keys proxy built in. Compare it one to one: boat vs Fly.io Machines and boat vs Agent 37.

Questions people ask

Is Fly.io Machines cheaper than Agent 37?

No, Agent 37's bill is lower in all five scenarios in the table. One agent on workdays is $6 a month on Agent 37 and $15 on Fly, and 50 agents in parallel are $329 versus $487, with Agent 37 on its VM type because the fleet needs Docker. Fly's figures already use its cheapest rate, shared vCPUs on a 1-year reservation.

Which is faster, Fly.io Machines or Agent 37?

There is no comparable measurement for either one. Both share CPUs by default and sell dedicated cores at a higher rate. boat did 26.69 runs per second on the public StarSling Node.js benchmark at 4 vCPU / 8 GB.

Can Fly.io Machines or Agent 37 run Docker?

On Fly you can install and run Docker yourself, but there is no /dev/kvm for nested VMs. On Agent 37 only the VM type runs Docker, at 2.5x the default compute rate. On boat, Docker, systemd and /dev/kvm are ready in every sandbox.

Do they keep my files when a machine stops?

A Fly Machine keeps only what is on its one Volume; its root file system is ephemeral. Agent 37 keeps the disk and can keep memory on auto-sleep. boat keeps the whole disk, with snapshots every minute and on stop.

Where do Fly.io Machines and Agent 37 run?

Fly runs Machines in 17 regions on six continents. Agent 37 runs on dedicated servers in the United States. boat runs only in the EU, in Germany, Finland and France, with EU data residency.

Which is better as an AI agent sandbox?

Agent 37 is built for it, with agents ready and a chat API, while on Fly you build the image, browser and agent control yourself. boat also ships agents ready, on a full VM with Docker, a desktop and forks of running machines.

Try boat

Install the CLI, then run boat new.

7-day free trial