DigitalOcean vs Hetzner Cloud

Both rent plain Linux VMs that you set up and run yourself. DigitalOcean bills per second and adds GPUs, more regions and managed services, while Hetzner Cloud bills every started hour, has lower list prices on its cheapest servers, and does not allow nested virtualization.

The companies and products

DigitalOcean

DigitalOcean is a general-purpose cloud, and Droplets are its Linux virtual machines.

A Droplet is a Linux VM you rent by size, with root, a public IP, local SSD and a monthly transfer allowance. You pick an image, then install and run your own software on it, next to DigitalOcean products like Kubernetes, managed databases, load balancers and cloud firewalls. It is a server host, not an agent sandbox, so this page compares the two where they overlap: giving an agent its own Linux machine. DigitalOcean also sells two separate products in preview, Harness Runtime for agents and MicroVMs, which are not Droplets.

Founded: 2012
Product: Linux-based virtual machines on virtualized hardware
CPU: Basic plans share a hyper-thread with other Droplets; dedicated CPU plans get the full hyper-thread
Regions: 16 datacenters in 13 regions across the US, Canada, Europe, Asia and Australia
Billing: Per second with a 60 second minimum, capped at 672 hours a month; powered-off Droplets are still billed
GPUs: Nvidia H100, H200, B300, L40S and RTX Ada cards, and AMD MI300X, MI325X, MI350X and MI355X
Agent product: Harness Runtime runs agents such as Claude Code and Codex in separate microVM sandboxes, apart from Droplets

Hetzner Cloud

Hetzner is a German hosting company founded in 1997; Hetzner Cloud rents virtual servers, storage and networking by the hour.

A Hetzner Cloud server is a Linux VM picked from shared vCPU lines (cost-optimized CX and CAX, regular CPX) or dedicated vCPU (CCX), with local NVMe disk, a public IP and up to 20 TB of traffic in the EU. You add volumes, private networks, firewalls and load balancers, and run everything yourself through the Console, the Cloud API, the hcloud CLI or Terraform. It is general infrastructure, not an agent sandbox, so this page compares the two where they overlap: giving an agent its own Linux machine.

Founded: 1997, in Germany
Product: Virtual machines on Hetzner's physical servers, with root and your own OS image
Billing: Hourly with a monthly cap; every started hour is billed, and a server bills until you delete it, even when it is off
Pricing change: New prices for new orders and rescales since 15 June 2026; older servers keep their old price until rescaled
Locations: Falkenstein and Nuremberg (Germany), Helsinki (Finland), Ashburn and Hillsboro (US), Singapore
Stock: The cheapest CX and CAX servers show "This product is currently unavailable. Please check back later."
Certifications: ISO/IEC 27001:2022

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. “39th cheapest of 368” means 38 of the other 367 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	DigitalOcean: price rank	Hetzner Cloud: price rank	bochka: price rank
Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Both give an agent a full VM with root and Docker, but Hetzner Cloud cannot run nested VMs and a new DigitalOcean team gets only 3 Droplets at once; bochka has /dev/kvm, snapshots and coding agents built in.	134th of 367, with limits $1,187 a month	46th of 367, with limits $287 a month	3rd cheapest of 367 $634 a month
Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB	35th of 366, with limits $1,429 a month	94th of 366, with limits $325 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 DigitalOcean fits short runs better than Hetzner Cloud: it bills per second with a 60 second minimum, while Hetzner bills every started hour in full.	63rd of 345, with limits $60 a month	156th of 345, with limits $649 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 Both need a limits raise first: a new DigitalOcean team runs 3 Droplets at once and Hetzner reviews server limit requests by hand, while bochka allows 100 to 2,000 sandboxes by plan.	54th of 361, with limits $421 a month	46th of 361, with limits $314 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	31st of 360, with limits $1,786 a month	104th of 360, with limits $6,490 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	106th of 364, with limits $357 a month	51st cheapest of 364 $1,010 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 A ten-minute CI job is billed as a full hour on Hetzner Cloud and by the second on DigitalOcean; on both you run the runner software and the clean-up yourself.	146th of 377, with limits $171 a month	83rd cheapest of 377 $788 a month	4th cheapest of 377 $86 a month
24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB All three fit a 24/7 dev box: DigitalOcean and Hetzner Cloud are classic always-on servers with a monthly price cap, and bochka keeps the disk with snapshots every minute.	39th cheapest of 368 $48 a month	32nd cheapest of 368 $42 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 Neither DigitalOcean nor Hetzner Cloud ships a desktop, so you install the desktop, the browser and VNC yourself; bochka streams a 60fps desktop with Chromium and computer use.	117th of 369, with limits $263 a month	134th of 369, with limits $320 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	56th of 367, with limits $147 a month	227th of 367, with limits $312 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	88th of 346, with limits $175 a month	157th of 346, with limits $62 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	69th of 338, with limits $1,786 a month	45th of 338, with limits $1,248 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 Many short parallel machines suit DigitalOcean's per-second billing better than Hetzner's hourly rounding, once your team has a higher Droplet limit.	22nd of 360, with limits $199 a month	114th of 360, with limits $52 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 DigitalOcean rents GPUs here, with Nvidia and AMD GPU Droplets; Hetzner Cloud and bochka have none.	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	DigitalOcean	Hetzner Cloud	bochka
What you get	A general Linux VM (Droplet) you set up and run yourself	A general Linux VM you set up and run yourself	A sandbox API for agents, on full VMs
Isolation	Every Droplet is a full VM with its own kernel	Every server is a full VM with its own kernel	Every sandbox is a full VM with its own kernel
CPU speed (Node.js build-and-test, 4 vCPU / 8 GB)	No comparable measurement; Basic plans share hyper-threads with other Droplets	8.88 runs/s on a shared CX33; CPX and CCX were not measured	26.69 runs/s on dedicated AMD Ryzen 9 9950X (Zen 5) hosts
Docker and nested virtualization	Root and Docker; nested VMs are mentioned in the live migration docs but not listed as a feature	Root and Docker; nested virtualization is "not possible on cloud server"	Docker, root, systemd and /dev/kvm on every machine, documented
Billing unit	Per second with a 60 second minimum, capped at 672 hours a month	Every started hour billed in full, with a monthly cap	Per second, spent from the usage credit of a monthly plan from $20
Stopping	A powered-off Droplet is billed in full until you destroy it	A powered-off server is billed in full until you delete it	A stopped sandbox costs nothing and keeps its disk
Snapshots and backups	On-demand snapshots billed by size; paid scheduled backups from every 4 hours to weekly	Snapshots you take yourself, or daily backups in 7 slots for 20% of the server price	Automatic snapshots every minute and on stop
Fork	Take a live snapshot, then create new Droplets from it (disk, no memory)	Take a snapshot, then create a new server from it (disk, no memory; shutting down first is recommended)	Fork a running sandbox in one call (disk, no memory)
Machines at once	A new team gets 3 Droplets; 100 at Tier 5, unlocked by prepayment, paid invoices or a request	Per-project limits are not published; raised by a support request reviewed by hand	100 to 2,000 sandboxes by plan
Reaching the machine	Public IPv4 and IPv6 on any port; HTTPS needs your own certificates or a load balancer	Public IPv4 and IPv6 on any port; HTTPS needs your own certificates or a paid load balancer	HTTPS URLs for up to 50 ports, token-protected; raw TCP; own IP
Coding agents, desktop and keys	None on Droplets; the separate Harness Runtime preview runs agents in microVMs; secrets go on the server	None; you install and drive agents yourself, for example over SSH, and put secrets on the server	Claude Code, Codex, OpenCode and more preinstalled with a prompt, steer and events API; 60fps desktop with Chromium; AI credentials proxy for any HTTPS API
Cloud around the server	Managed databases, Kubernetes, load balancers, VPC and free firewalls with outbound rules	Volumes, private networks, load balancers and free firewalls with outbound rules	A sandbox product; open internet on every plan, no provider-side egress allowlist
GPUs	Nvidia and AMD GPU Droplets, on demand, reserved or spot	None in Hetzner Cloud	None
Regions	13 regions in the US, Canada, Europe, Asia and Australia	Germany and Finland, plus Ashburn, Hillsboro and Singapore	EU only: Germany, Finland, France

Every feature Battleships tracks (43)

Feature	DigitalOcean	Hetzner Cloud	bochka
Full VM (own kernel)	Yes	Yes	Yes
Docker inside	Yes	Yes	Yes
Nested virtualization	Yes	No	Yes
Root	Yes	Yes	Yes
Systemd	Yes	Yes	Yes
Desktop GUI	No	No	Yes
Browser	No	No	Yes
Computer-use API	No	No	Yes
SSH	Yes	Yes	Yes
Persistent disk	Yes	Yes	Yes
Pause/resume	No	Yes	Yes
Automatic snapshots	No	No	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	No	No	No
Wake on request	No	No	No
HTTPS preview URLs	No	No	Yes
Raw TCP inbound	Yes	Yes	Yes
Public IPv4	Yes	Yes	No
Egress allowlist	Yes	Yes	No
Secret proxy	No	No	Yes
Secret proxy for any API	No	No	Yes
Preinstalled agents	No	No	Yes
Agent harness API	No	No	Yes
MCP server	Yes	No	No
GPU	Yes	No	No
Windows	No	No	No
Arm64	No	Yes	No
Self-hosting	No	No	No
BYOC	No	No	No
Open source	No	No	No
Scoped API keys	Not published	Not published	Yes
Spend limits	No	No	Yes
Usage per sandbox (API)	Yes	No	Yes
Audit logs	Not published	Not published	No
SSO	Yes	No	No
SOC 2	Yes	No	No
HIPAA	Yes	No	No
EU data residency	Yes	Yes	Yes
Zero data retention	Not published	Not published	Yes

Performance

DigitalOceanNot measured

Hetzner Cloud8.88 runs/s

bochka26.69 runs/s

In StarSling's public benchmark at 4 vCPU / 8 GB, bochka did 26.69 Node.js build-and-test runs per second on its own hosts. A shared Hetzner CX33 did 8.88 runs per second in an earlier run, about a third of bochka's speed; Hetzner's CPX and dedicated CCX servers were not measured. DigitalOcean Droplets have no comparable measurement.

Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Hetzner Cloud measured on: Shared vCPU, Cost-Optimized (CX, Intel/AMD), EU. DigitalOcean has 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	DigitalOcean, a month	Hetzner Cloud, 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	$13 · Pay as you go	$3 · Pay-as-you-go (postpaid) · with limits	$20 · $20/mo
One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month	$48 · Pay as you go	$10 · Pay-as-you-go (postpaid) · 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	$629 · Pay as you go · with limits	$141 · Pay-as-you-go (postpaid) · 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	$179 · Pay as you go · with limits	$312 · Pay-as-you-go (postpaid) · 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	$357 · Pay as you go · with limits	$2,080 · Pay-as-you-go (postpaid) · with limits	$2,000 · $2000/mo · with limits

DigitalOcean: Each Droplet size has a fixed hourly price, billed per second with a 60 second minimum and capped at 672 hours a month. Billing runs from create to destroy, including while the Droplet is powered off, and snapshots and backups cost extra.

Hetzner Cloud: Hetzner bills each server by the hour, rounding every started hour up, with a monthly cap, until you delete it, even when it is off. The prices here use the cheapest shared CX servers, which are currently out of stock; the CPX and CCX servers you can order today cost more per hour.

Hetzner Cloud: 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

A new DigitalOcean team can run only 3 Droplets at once, Basic shared plans up to $48, so 50 agents in parallel or sandboxes for your users need a higher tier first; Tier 5 allows 100. Hetzner does not publish its server limits, reviews each raise by hand, and every scenario here is priced on its CX line, which Hetzner marks as currently unavailable. Short jobs on Hetzner are also billed a full hour each. On bochka the limit is 100 to 2,000 sandboxes by plan plus starts per minute, so the 1,000-job burst is queued and does not fit either.

Scenario	DigitalOcean	Hetzner Cloud	bochka
One coding agent, workdays	None that bite	stock-limited: all CX types "currently unavailable" on hetzner.com (2,026-09-28)	None that bite
One agent that never stops	None that bite	stock-limited: all CX types "currently unavailable" on hetzner.com (2,026-09-28)	None that bite
50 agents in parallel	Pay as you go: max 3 concurrent	stock-limited: all CX types "currently unavailable" on hetzner.com (2,026-09-28)	None that bite
Sandboxes for your users	Pay as you go: max 3 concurrent	stock-limited: all CX types "currently unavailable" on hetzner.com (2,026-09-28)	None that bite
1,000 short jobs at once	Pay as you go: max 3 concurrent shared vCPUs on this tier, per-second with 672 h cap”): heavy, sustained CPU use can be slowed by other tenants	stock-limited: all CX types "currently unavailable" on hetzner.com (2,026-09-28)	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 DigitalOcean when

you want per-second billing, GPUs, regions in North America, Asia or Australia, or servers next to managed databases and Kubernetes.

Per-second billing. Useful for short or bursty jobs: Droplets bill per second with a 60 second minimum, where Hetzner rounds every started hour up.
GPUs and more regions. Useful if you need Nvidia or AMD GPUs, or servers in Canada, India or Australia, which Hetzner Cloud does not offer.
Managed services next to the server. Useful if the machine must sit beside managed databases and Kubernetes clusters in one private network and one account.

Use Hetzner Cloud when

you want low-priced long-lived servers in Europe, big dedicated machines or lots of included traffic, and do not need nested virtualization.

Lower list prices on the cheapest servers. Useful for long-running servers in Europe, as long as the CX line, currently out of stock, is back.
Big dedicated machines. Useful for heavy single machines: dedicated CCX servers go up to 48 vCPU and 192 GB of memory.
Lots of included traffic in the EU. Useful for data-heavy servers: each EU server includes 20 TB of outbound traffic, and extra traffic costs $1.20 per TB.

Where bochka fits

DigitalOcean and Hetzner Cloud are general clouds, and bochka is not trying to be one: it has no GPUs, no managed databases and runs only in the EU. If what you are really running is agents, each needing its own Linux machine that starts ready, pauses for free and can be copied, bochka does that part for you on full VMs.

Free stops and a disk saved every minute

Both clouds bill a powered-off machine in full, so an idle agent keeps costing money. The workaround is to snapshot, delete and recreate by script, losing changes since the last snapshot. bochka saves every minute and stops for free.

Agents, a desktop and a key proxy built in

Droplets and Hetzner servers start as empty OS images, so you install, update and supervise each agent, desktop and API key yourself, and the agent can read any key you put there. bochka ships them and keeps keys outside.

Nested virtualization and a faster CPU

Hetzner Cloud cannot run VMs inside a server, and DigitalOcean does not list nested virtualization as a feature. bochka documents /dev/kvm on every machine, and measured about three times the CPU speed of a Hetzner CX33.

Use bochka when the machines are for agents: a full VM each with /dev/kvm, coding agents, a desktop and HTTPS ports built in, snapshots every minute, fork while running, and stops that cost nothing. Compare it one to one: bochka vs DigitalOcean and bochka vs Hetzner Cloud.

Questions people ask

Is DigitalOcean cheaper than Hetzner Cloud?

On list prices for long-running servers, no: one agent that never stops comes to $48 a month on DigitalOcean against $10 on Hetzner's CX servers, and bochka comes to $26. But the CX line is currently out of stock, and for 300 short user sandboxes Hetzner's hourly rounding brings it to $312 against $179 on DigitalOcean and $200 on bochka.

Which is faster, DigitalOcean or Hetzner Cloud?

There is no comparable measurement for DigitalOcean Droplets. A shared Hetzner CX33 did 8.88 Node.js build-and-test runs per second at 4 vCPU / 8 GB, against 26.69 for bochka on dedicated AMD Ryzen 9 9950X hosts.

Can DigitalOcean and Hetzner Cloud run Docker and nested VMs?

Both run Docker, since each machine is a full VM with root. Hetzner's server FAQ says nested virtualization is not possible on cloud servers, and DigitalOcean mentions nested VMs only in its live migration docs, not as a listed feature. bochka documents Docker, systemd and /dev/kvm on every machine.

Do I pay for a powered-off server on DigitalOcean or Hetzner?

Yes, on both. A Droplet or a Hetzner server is billed in full until you destroy or delete it, so to stop paying you snapshot it, delete it and create it again later. On bochka, a stopped sandbox costs nothing and keeps its disk.

Which has more regions and GPUs?

DigitalOcean: 13 regions across the US, Canada, Europe, Asia and Australia, and Nvidia and AMD GPU Droplets. Hetzner Cloud runs in Germany, Finland, Ashburn, Hillsboro and Singapore, with no GPUs. bochka runs only in the EU and has no GPUs.

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