Nebius vs Cerebrium
Nebius is an AI cloud that rents you full GPU and CPU virtual machines, with root and SSH, plus clusters and storage around them. Cerebrium is a serverless GPU cloud: you hand it your code or Dockerfile and it serves your app as containers that scale up with requests and back down to zero.
The companies and products
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.
Cerebrium
Cerebrium is a serverless GPU cloud for real-time AI: voice agents, video models, LLMs and other ML apps.
You describe an app in a cerebrium.toml file, with your Python code or your own Dockerfile, and Cerebrium builds it and serves it as HTTP or WebSocket endpoints. It starts containers when requests arrive, scales them across GPUs and regions, and scales them back to zero when traffic stops. It is built for teams that serve models in production, not for agents that need a machine of their own, so this page compares the places where the two overlap.
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. “29th cheapest of 364” means 28 of the other 363 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 | Nebius: price rank | Cerebrium: 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, root and a disk that lasts; Nebius gives you a clean Ubuntu VM to set up, Cerebrium keeps only one folder, and boat ships agents ready. | 197th of 367, with limits $2,272 a month | 182nd of 367, with limits $2,008 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB | 210th of 366, with limits $2,261 a month | 221st of 366, with limits $2,534 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 Short code runs suit Cerebrium, whose containers start on a request and scale back to zero; it comes to $131 a month against $202 on Nebius. | 111th of 345, with limits $202 a month | 225th of 345, with limits $131 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 300 user sandboxes at once need a quota raise on Nebius past 12 VMs per region, Cerebrium's Hobby plan allows 500 CPU containers, and boat gives each user a full VM with hosting. | 81st of 361, with limits $716 a month | 230th of 361, with limits $821 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 | 52nd of 360, with limits $3,049 a month | 72nd of 360, with limits $4,068 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 All three fit agent evals: Nebius comes to $610 a month, Cerebrium to $639 a month and boat to $200 a month. | 29th cheapest of 364 $610 a month | 32nd cheapest of 364 $639 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 Both fit as CI runners: Nebius comes to $297 a month against $766 on Cerebrium, and boat adds Docker and /dev/kvm on every machine for $86 a month. | 27th cheapest of 377 $297 a month | 81st cheapest of 377 $766 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB A Nebius VM gives you SSH and a disk that keeps your installs, while Cerebrium has no SSH and resets everything outside /persistent-storage. | 68th cheapest of 368 $91 a month | 259th of 368, with limits $102 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 Nebius nor Cerebrium has a desktop or browser product, while boat streams a 60fps desktop with Chromium on every sandbox. | 187th of 369, with limits $610 a month | 179th of 369, with limits $575 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 | 223rd of 367, with limits $305 a month | 221st of 367, with limits $287 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 | 253rd of 346, with limits $302 a month | 223rd of 346, with limits $195 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 | 174th of 338, with limits $6,049 a month | 135th of 338, with limits $3,991 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 | 176th of 360, with limits $156 a month | 197th of 360, with limits $191 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 Both fit GPU eval jobs: Nebius comes to $1,127 a month and Cerebrium to $1,975 a month, and boat has no GPUs. | 18th cheapest of 98 $1,127 a month | 28th cheapest of 98 $1,975 a month | 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 | Nebius | Cerebrium | boat |
|---|---|---|---|
| Kind of product | AI cloud: GPU and CPU VMs, Kubernetes, Slurm clusters, serverless jobs and storage | Serverless GPU inference: your app runs as autoscaling containers behind endpoints | Sandboxes for AI agents: VMs with agents, desktop and hosting ready |
| What you get and how you reach it | A full VM behind a hypervisor with its own kernel, root and SSH | Containers built from your code or Dockerfile; no SSH, you reach the app through HTTP or WebSocket endpoints | A full VM with its own kernel, root, SSH and systemd, on dedicated AMD Ryzen 9 9950X hosts |
| Docker and nested virtualization | Root on your own VM and container VMs from Docker images; nested virtualization is not documented | The app is the container; Docker inside and /dev/kvm are not documented | Docker, root, systemd and /dev/kvm on every machine |
| GPUs | Nvidia L40S, RTX PRO 6000, H100, H200, B200 and B300, on-demand or preemptible | T4, A10, L4, L40S, A100, H100, H200, B200 and RTX PRO 6000, up to 8 per container, plus Inferentia2 and Trainium | None, CPU only |
| Biggest machine | CPU VMs up to 256 vCPU / 1,024 GB | Up to 24 vCPU / 256 GB on H100, H200 and B200 containers, 48 vCPU / 192 GB on L4 and T4 | 16 vCPU / 32 GB on the $100 plan and up |
| Scale to zero and wake | No; you start and stop VMs yourself | Yes; apps scale to zero and start again on the next request, up to your replica limit | No; a stopped sandbox is started by API |
| Files after a stop | Disks are kept, and keep billing on their full provisioned size while the VM is stopped | Only the shared /persistent-storage volume survives; packages and files elsewhere reset | The whole disk is saved every minute and on stop, and a stopped sandbox keeps it for free |
| Snapshots and fork | Manual disk snapshots; to fork, make a disk from the snapshot, then a VM from the disk | No fork; an early-beta memory checkpoint (no files) speeds up new containers of the same build | Snapshots every minute and on stop; fork a running sandbox in one call |
| Run time | Regular VMs run until you stop them; preemptible VMs can be stopped at any time | No container cap, but a request times out after 15 minutes by default, and async runs stop at 12 hours | No session cap on paid plans; you set the time to live |
| Inbound access | Optional public IPv4 per VM, SSH and Nebius Tunnels; you set up TLS yourself | One HTTP or WebSocket endpoint per app on one port; custom domains on Standard | HTTPS URLs for up to 50 ports, token-protected; raw TCP; dedicated IPv4 or IPv6 |
| Network rules | VPC security groups allow or deny ingress and egress by IP range, port and protocol | No egress allowlist documented | Open internet on every plan; no provider-side allowlist |
| Desktop and browser | No desktop or browser product; a text serial console | No desktop or browser product | 60fps desktop with Chromium, computer use and browser control, embeddable stream |
| Coding agents and keys | No agents built in; SecretStash hands secrets to the VM itself | No agents to run inside; secrets arrive in the app as environment variables | Claude Code, Codex, OpenCode and more preinstalled with a prompt, steer and events API; a credentials proxy keeps real keys outside |
| Regions | Finland, France, Spain, United Kingdom, Israel and two US regions | US East, US Central, Stockholm and Finland; Tokyo, Mumbai, Sao Paulo and 5 more on request | EU only: Germany, Finland, France |
Every feature Battleships tracks (43)
| Feature | Nebius | Cerebrium | boat |
|---|---|---|---|
| Full VM (own kernel) | Yes | No | Yes |
| Docker inside | Yes | No | Yes |
| Nested virtualization | Not published | No | Yes |
| Root | Yes | Not published | Yes |
| Systemd | Not published | Not published | Yes |
| Desktop GUI | No | No | Yes |
| Browser | No | No | Yes |
| Computer-use API | No | No | Yes |
| SSH | Yes | No | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | No | Yes |
| Automatic snapshots | No | No | Yes |
| Fork/clone | Yes | No | Yes |
| Live fork (no pause) | Yes | No | Yes |
| Memory fork | No | No | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | No | Yes | No |
| Wake on request | No | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | Not published | Not published | Yes |
| Public IPv4 | Yes | No | No |
| Egress allowlist | Yes | No | 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 | No | No | No |
| GPU | Yes | Yes | No |
| Windows | No | No | No |
| Arm64 | No | No | 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 | Not published | Not published | Yes |
| Usage per sandbox (API) | Yes | Yes | Yes |
| Audit logs | Not published | Yes | No |
| SSO | Yes | No | No |
| SOC 2 | Yes | Yes | No |
| HIPAA | Yes | Yes | No |
| EU data residency | Yes | Yes | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
There is no comparable measurement for either: neither Nebius nor Cerebrium was part of the public StarSling benchmark run, so we cannot say which is faster at CPU work. 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 hosts. Nebius CPU-only VMs run on AMD EPYC Genoa or Intel Ice Lake, and Cerebrium's own speed claims are about GPU inference and cold starts.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Nebius and Cerebrium 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 | Nebius, a month | Cerebrium, 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 | $25 · Pay as you go | $53 · Hobby · with limits | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $91 · Pay as you go | $210 · Hobby · 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 | $1,127 · Pay as you go | $2,557 · Hobby · 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 | $305 · Pay as you go | $715 · Hobby | $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 | $610 · Pay as you go | $1,681 · Standard | $2,000 · $2000/mo · with limits |
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 tables use the rates from 1 October 2026: AMD Genoa costs $0.015 per vCPU-hour plus $0.0045 per GiB-hour.
Cerebrium: Cerebrium bills per second for each vCPU, GB of memory and GPU while a container runs. Hobby has no fee; Standard is $100 a month on top of compute with no usage credit, the protected tier costs 2x on all compute, and storage past the first 100 GB costs $0.05 per GB a month.
Cerebrium: Plans start at $20 a month, and the whole fee is usage credit spent per second on running sandboxes. A stopped sandbox costs nothing and keeps its disk.
Limits in those scenarios
Cerebrium does not fit the coding agent, always-on and 50-agent scenarios: only /persistent-storage survives a stop, so installed packages reset, and the 50-agent fleet also needs Docker inside, which is not documented. Cerebrium also does not publish an egress price, so the scenarios count none. Nebius fits all five, but it starts each region at 12 VMs and, in most regions, 200 vCPU across CPU-only VMs, so 50 agents, 300 user sandboxes and 1,000 short jobs at once need a quota raise first. boat's start limits queue the 1,000-job burst, even on its $2,000 plan.
| Scenario | Nebius | Cerebrium | boat |
|---|---|---|---|
| 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
you want full VMs you control, the biggest GPUs, a Slurm or Kubernetes cluster, a very large CPU machine, or a region in the UK, Israel or the US.
- A real VM that stays up. Useful when you want a plain Linux VM with root and SSH that runs until you stop it and keeps its disks.
- The biggest GPUs and very large machines. Useful for training or one huge job: Nebius rents Nvidia B300 GPUs and CPU VMs up to 256 vCPU / 1,024 GB.
- A full cloud around the machine. Useful when you need egress rules by IP range, private networks, Kubernetes, Slurm or object storage next to your VMs.
Use Cerebrium when
you serve a model, voice or video app as an endpoint that should scale to zero between requests and grow with traffic.
- Endpoints that scale to zero. Useful for apps that wait for traffic: they scale to zero, start on the next request and grow under load.
- A wide range of GPUs per container. Useful for serving models: GPUs from T4 to B200, up to 8 in one container, plus AWS Inferentia2 and Trainium.
- Regions on request. Useful for users far from the EU and US: support can open Tokyo, Mumbai, Sao Paulo, Dubai and more.
Where boat fits
Nebius and Cerebrium are GPU clouds, and boat has no GPUs, so model training and serving belong on one of them. If what you are really running is agents, boat gives each one a full VM that comes with coding agents, a desktop and a disk saved every minute. You can keep the GPU work where it is and run the agents on boat.
A whole disk that stays, free while stopped
On Cerebrium an agent loses every package it installed outside /persistent-storage when the container scales down. On Nebius the disk stays, but it keeps billing while stopped and you snapshot by hand.
Agents and a desktop ready on every machine
Neither has an agent harness or a desktop. On Nebius you install the agent, a desktop, a VNC server and a browser on each VM yourself; on Cerebrium there is no shell to install them in.
Real keys kept out of the machine
Nebius SecretStash hands the secret to the VM and Cerebrium puts secrets in environment variables, so a prompt-injected agent can read and leak them. boat's credentials proxy keeps keys outside, for any HTTPS API.
Use boat when you run agents: full VMs that come with coding agents, a desktop and browser, HTTPS ports, a credentials proxy and a disk saved every minute, free while stopped. Compare it one to one: boat vs Nebius and boat vs Cerebrium.
Questions people ask
Is Nebius cheaper than Cerebrium?
In all five scenarios we priced, Nebius came out lower. One agent that never stops is $91 against $210 a month, 50 agents in parallel are $1,127 against $2,557 a month, and 1,000 short jobs at once are $610 against $1,681 a month. On boat the always-on agent is $26 a month.
Which is faster, Nebius or Cerebrium?
We have no comparable CPU measurement for either, since neither was part of the public StarSling benchmark run. boat did 26.69 runs per second at 4 vCPU / 8 GB on that test.
Can I run Docker on Nebius or Cerebrium?
On Nebius you get root on your own VM, so you can install Docker, and it can also start container VMs from Docker images. Cerebrium builds your app from a Dockerfile, but running Docker inside the app is not documented. A boat sandbox runs Docker, systemd and /dev/kvm like a normal Ubuntu server.
Which has better GPUs, Nebius or Cerebrium?
Nebius goes up to Nvidia B300 and adds Slurm and Kubernetes clusters for training. Cerebrium covers smaller GPUs like T4, A10 and L4 too, up to 8 per container, plus Inferentia2 and Trainium. For GPU eval jobs Nebius comes to $1,127 a month and Cerebrium to $1,975 a month; boat has no GPUs.
What happens to my files when a machine stops?
A stopped Nebius VM keeps its disks, which keep billing on their full size. A Cerebrium container keeps only the /persistent-storage volume and resets the rest. A stopped boat sandbox keeps its whole disk for free.
Try boat
Install the CLI, then run boat new.
7-day free trial