boat vs Nebius
Nebius is an AI cloud: you rent Nvidia GPUs, plain CPU virtual machines, Kubernetes and Slurm clusters, and build the rest yourself. boat is built for agents, with coding agents you drive by API, a 60fps desktop and snapshots every minute on every VM.
Where boat stands out
Coding agents built in, keys kept outside
boat ships Claude Code, Codex, OpenCode and other agents inside each sandbox, and you drive them from outside with a prompt, steer and events API. boat's credentials proxy keeps real AI keys out of the sandbox and can switch to another account when one runs out of credits. A Nebius VM is a clean Ubuntu machine, and SecretStash hands the secret to the VM itself.
A desktop and a browser on every sandbox
Every boat sandbox has a 60fps desktop with Chromium, computer use and browser control, and you can embed the stream in your own app. Nebius has no desktop or browser product: a VM gives you SSH and a text serial console.
Snapshots every minute, fork in one call
boat snapshots each sandbox's disk every minute and on stop, and you can fork a running sandbox in one call. A stopped sandbox costs nothing. On Nebius you take disk snapshots by hand, and a fork means making a new disk from the snapshot and then a new VM from that disk.
The companies and products
boat
boat.dev runs sandboxes for AI agents: each one is a full Linux VM with its own kernel.
You start a machine with one command or one API call, then SSH in, run Docker, open a desktop, or let a built-in agent (Claude Code, Codex, OpenCode and others) work in it. The disk stays between sessions, and a stopped sandbox costs nothing.
- Isolation
- Full VM per sandbox, own kernel
- Regions
- EU: Germany, Finland, France
Nebius
Nebius is an AI cloud company, listed on Nasdaq and based in Amsterdam, that rents GPUs and the infrastructure around them.
You create a Compute VM from a preset (GPU or CPU only), add disks and an SSH key, and get a full Linux VM with root and an optional public IP. Around it Nebius sells Managed Kubernetes, Slurm clusters (Soperator), Serverless AI jobs and endpoints, object storage and networking. It is built for teams that train, fine-tune and serve AI models, not as a sandbox for agents.
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. “3rd cheapest of 367” means 2 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 | boat: price rank | Nebius: price rank |
|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A Nebius VM can run a coding agent, but you install the agent, Docker and the tooling yourself and fork by hand through disk snapshots. | 3rd cheapest of 367 $634 a month | 197th of 367, with limits $2,272 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An always-on assistant fits on a small Nebius CPU VM, but the agent, browser and desktop are yours to set up; boat ships them ready. | 5th cheapest of 366 $657 a month | 210th of 366, with limits $2,261 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month | 186th of 345, with limits $2,000 a month | 111th of 345, with limits $202 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month 300 user sandboxes on Nebius start far above its default quota of 12 VMs per region, so you need a quota raise first. | Cheapest of 361 $209 a month | 81st of 361, with limits $716 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 34th of 360, with limits $2,000 a month | 52nd of 360, with limits $3,049 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 10th cheapest of 364 $200 a month | 29th cheapest of 364 $610 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Both fit: a Nebius CPU VM works as a CI runner, and boat adds snapshots every minute and free stops between jobs. | 4th cheapest of 377 $86 a month | 27th cheapest of 377 $297 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB | 18th cheapest of 368 $26 a month | 68th cheapest of 368 $91 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Nebius has no desktop or computer-use product, only a text serial console, while boat streams a 60fps desktop with Chromium. | 6th cheapest of 369 $200 a month | 187th of 369, with limits $610 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month On Nebius you install and run the browser yourself on a plain VM; boat has Chromium and browser control on every sandbox. | 7th cheapest of 367 $200 a month | 223rd of 367, with limits $305 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 6th cheapest of 346 $90 a month | 253rd of 346, with limits $302 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 71st of 338, with limits $1,800 a month | 174th of 338, with limits $6,049 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month | 2nd cheapest of 360 $200 a month | 176th of 360, with limits $156 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month Only Nebius fits: it rents Nvidia GPUs from L40S to B300, and boat has no GPUs. | Can't run it | 18th cheapest of 98 $1,127 a month |
| 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 |
Key features
| Feature | boat | Nebius |
|---|---|---|
| Kind of product | Sandboxes for AI agents: VMs with agents, desktop and hosting ready | AI cloud: GPU and CPU VMs, Kubernetes, Slurm clusters and serverless jobs |
| Isolation | Every sandbox is a full VM with its own kernel | Every Compute VM is a full VM behind a hypervisor |
| Docker and nested virtualization | Docker, root, systemd and /dev/kvm on every machine | Root on your own VM and container VMs from Docker images; nested virtualization is not documented |
| GPUs | None | Nvidia L40S, RTX PRO 6000, H100, H200, B200 and B300, on-demand or preemptible |
| Biggest machine | 16 vCPU / 32 GB on the $100 plan and up | CPU VMs up to 256 vCPU / 1,024 GB |
| Stopping a machine | A stopped sandbox costs nothing and keeps its disk | Compute stops billing; disks keep billing on their full provisioned size |
| Snapshots and fork | Snapshots every minute and on stop; fork a running sandbox in one call | Manual disk snapshots of a running VM; to fork, make a disk from the snapshot, then a VM from the disk |
| Desktop and computer use | 60fps desktop and Chromium, computer use and browser control, embeddable stream | No desktop or browser product; a text serial console |
| Coding agents | Claude Code, Codex, OpenCode and more preinstalled, with a prompt, steer and events API | None built in; you install and run agents yourself over SSH |
| Reaching the machine | HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv6 or IPv4 | Optional public IPv4 per VM, SSH and Nebius Tunnels; you set up TLS yourself |
| Network rules | Open internet on every plan; no provider-side allowlist | VPC security groups allow or deny ingress and egress by IP range, port and protocol |
| Keeping keys out of the machine | AI credentials proxy for model providers and any HTTPS API you register | SecretStash stores secrets that a VM fetches with its service account |
| Regions | EU only: Germany, Finland, France | Finland, France, Spain, United Kingdom, Israel and the US |
| Compliance and transparency | SOC 2 is in progress. We're open about where we stand: live status and data retention policy. | SOC 2 |
Every feature Battleships tracks (43)
| Feature | boat | Nebius |
|---|---|---|
| Full VM (own kernel) | Yes | Yes |
| Docker inside | Yes | Yes |
| Nested virtualization | Yes | Not published |
| Root | Yes | Yes |
| Systemd | Yes | Not published |
| Desktop GUI | Yes | No |
| Browser | Yes | No |
| Computer-use API | Yes | No |
| SSH | Yes | Yes |
| Persistent disk | Yes | Yes |
| Pause/resume | Yes | Yes |
| Automatic snapshots | Yes | No |
| Fork/clone | Yes | Yes |
| Live fork (no pause) | Yes | Yes |
| Memory fork | No | No |
| Custom image (Docker or snapshot) | Yes | Yes |
| ≥24 h sessions | Yes | Yes |
| Idle auto-stop | No | No |
| Wake on request | No | No |
| HTTPS preview URLs | Yes | Yes |
| Raw TCP inbound | Yes | Not published |
| Public IPv4 | No | Yes |
| Egress allowlist | No | Yes |
| Secret proxy | Yes | No |
| Secret proxy for any API | Yes | No |
| Preinstalled agents | Yes | No |
| Agent harness API | Yes | No |
| MCP server | No | No |
| GPU | No | Yes |
| Windows | No | No |
| Arm64 | No | No |
| Self-hosting | No | No |
| BYOC | No | No |
| Open source | No | No |
| Scoped API keys | Yes | Not published |
| Spend limits | Yes | Not published |
| Usage per sandbox (API) | Yes | Yes |
| Audit logs | No | Not published |
| SSO | No | Yes |
| SOC 2 | No | Yes |
| HIPAA | No | Yes |
| EU data residency | Yes | Yes |
| Zero data retention | Yes | Not published |
Performance
There is no comparable measurement: Nebius was not part of the public StarSling benchmark run. boat did 26.69 Node.js build-and-test runs per second at 4 vCPU / 8 GB in that run, on dedicated AMD Ryzen 9 9950X (Zen 5) hosts. Nebius CPU-only VMs run on AMD EPYC Genoa or Intel Ice Lake.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Nebius 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 | boat, a month | Nebius, 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 | $20 · $20/mo | $25 · Pay as you go |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $26 · $20/mo | $91 · Pay as you go |
| 50 agents in parallel 50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside | $317 · $20/mo × 3 seats | $1,127 · Pay as you go |
| Sandboxes for your users 2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once | $200 · $100/mo × 2 seats | $305 · Pay as you go |
| 1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month | $2,000 · $2000/mo · with limits | $610 · Pay as you go |
boat: 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.
Nebius: Nebius bills running VMs per second for vCPU and RAM, with no plan fee, and bills disks by provisioned size even while the VM is stopped. The table uses the rates from 1 October 2026: AMD Genoa costs $0.015 per vCPU-hour plus $0.0045 per GiB-hour.
Limits in those scenarios
Nebius starts each region at 12 VMs and, in most regions, 200 vCPU across CPU-only VMs (0 in eu-north1 in Finland and us-central1, so a CPU VM there needs a raise first), and you raise that by asking. So 50 agents in parallel, 300 sandboxes for your users and 1,000 short jobs at once all need a bigger quota before they run. Its smallest CPU VM is 2 vCPU / 8 GB, so small sandboxes carry more memory than they need. On boat the limit is a sandbox count by plan plus starts per minute, so the 1,000-job burst is queued and needs more starts per day than the $2,000 plan allows.
| Scenario | boat | Nebius |
|---|---|---|
| One coding agent, workdays | None that bite | None that bite |
| One agent that never stops | None that bite | 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 |
| None that bite |
Which one should you use?
Use Nebius when
your work needs GPUs, a Slurm or Kubernetes cluster, a single very large machine, or a region in the UK, Israel or the US.
- GPUs and GPU clusters. Useful if your work trains, fine-tunes or serves models: Nebius rents Nvidia GPUs from L40S to B300, on-demand or preemptible.
- Very large machines and more regions. Useful for one huge job or a region outside the EU: Nebius CPU VMs reach 256 vCPU and run in the UK, Israel and the US.
- Network rules and a full cloud around the VM. Useful when you need egress rules by IP range, private networks, Kubernetes or object storage next to your machines.
Use boat when
you run agents: VMs that come with coding agents, a desktop and browser, HTTPS ports, a credentials proxy and snapshots every minute, and that cost nothing while stopped.
- Coding agents built in, keys kept outside. Without these, you write the setup scripts and the runner that starts, prompts and watches each agent. On Nebius the real key ends up on the machine, where an agent can print it to a log or commit it.
- A desktop and a browser on every sandbox. A computer-use or browser agent needs a screen to look at. On Nebius you install a desktop, a VNC server and a browser on each VM and keep them working yourself.
- Snapshots every minute, fork in one call. Agents break things and you want to roll back or branch. On Nebius there is no schedule, so the last good state is whenever you last snapshotted, and every stopped VM still pays for its disks.
Questions people ask
Is boat cheaper than Nebius?
In most of the scenarios we priced, yes. One coding agent on workdays costs $20 a month on boat and $25 on Nebius, one agent running all month is $26 against $91 a month, and 50 agents in parallel are $317 against $1,127. For 1,000 short jobs at once Nebius comes to $610 and boat to $2,000, and boat's start limits do not fit that burst.
Is Nebius a sandbox for AI agents?
No. Nebius is an AI cloud that rents GPU and CPU virtual machines, clusters and storage. You get a full Linux VM with root, but the agent, the desktop, the browser and the snapshot schedule are yours to build. boat ships those ready on every sandbox.
Does boat have GPUs like Nebius?
No. boat runs Linux VMs on dedicated AMD Ryzen 9 9950X hosts and has no GPUs. If your agents need to train or serve models, rent the GPUs on Nebius.
Does a stopped Nebius VM cost money?
Partly. Nebius stops charging for vCPU, RAM and GPUs when a VM is stopped, but its disks keep billing on their full provisioned size. A stopped boat sandbox costs nothing and keeps its disk.
Where does my data run?
boat runs only in the EU (Germany, Finland and France), with EU data residency and a zero data retention option. Nebius has EU regions in Finland, France and Spain, plus the United Kingdom, Israel and the US.
Is boat SOC 2 compliant?
SOC 2 is in progress. We're open about where we stand: live status at status.ascii.dev and our data retention policy at docs.boat.dev/data-retention.
Start a sandbox
Install the CLI, then run boat new.
7-day free trial
More comparisons
- boat vs Verda
- boat vs Jarvislabs
- boat vs Cerebrium
- boat vs Runpod
- boat vs Crusoe
- boat vs Paperspace
- boat vs Replicate
- boat vs SF Compute
- boat vs E2B
- boat vs Daytona
- boat vs Modal
- boat vs Vercel Sandbox
- Runpod vs Nebius
- Nebius vs Crusoe
- Nebius vs Paperspace
- Nebius vs Jarvislabs
- Nebius vs Hyperstack
- Nebius vs Verda
- Nebius vs Cerebrium
- Nebius vs Baseten