۵۰٪ تخفیف روی همه پلان‌ها، محدود. شروع از $2.48/mo

Jupyter Notebook VPS Hosting

Jupyter, in the cloud.
Run notebooks from any browser.

JupyterLab on AMD EPYC and pure NVMe, in any region.
Independent since 2008. SSH, conda, pip, GPU-ready, your own kernels.

4.6 · ۷۰۸ نظر در Trustpilot
Deploy Jupyter VPS Compare plans

From $7.48/mo · 50% off · No credit card required

~ ssh root@vps-fra-001 connected
root@vps-fra-001:~# conda install -c conda-forge jupyterlab
Solving environment ... done.
Installed 245 packages successfully.
root@vps-fra-001:~# jupyter lab --no-browser --port=8888 --ip=0.0.0.0
Jupyter Server 2.x is running at:
http://0.0.0.0:8888/lab?token=********
root@vps-fra-001:~# systemctl enable jupyterlab
Created symlink /etc/systemd/system/jupyterlab.service.
root@vps-fra-001:~# _

Jupyter VPS at a glance

Cloudzy hosts Jupyter Notebook VPS instances from 12 regions worldwide, starting at $7.48 / month. Each VPS runs on AMD EPYC با DDR5 memory, فضای ذخیره‌سازی NVMe, and a 40 Gbps uplink. Run JupyterLab or classic Jupyter behind Nginx with free TLS. GPU plans available for ML training. Cloudzy has been independent since 2008, serves 122,000+ developers, and is rated 4.6/5 by 679+ reviewers on Trustpilot.

Starting price
$7.48 / month
Provisioning
60 seconds
منطقه‌ها
۱۲ منطقه در سراسر جهان
آپتایم SLA
99.95%
بازگشت وجه
14 days
تأسیس
2008

Why builders pick Cloudzy

A Jupyter host that doesn't kick you off mid-cell.

NVMe + DDR5

Pure NVMe storage and DDR5 memory on AMD EPYC. Workloads stay fast under load.

Root SSH access

Full control of your stack. Install packages, tune configs, run cron jobs, deploy via git. No restrictions.

12 regions

Deploy as close to your users as physics allows. Median P50 latency under 10 ms in NA/EU.

Real human support

Chat with engineers who know your stack, not script-readers. Median resolution under 1 hour.

Stack you'll run

Every Jupyter dependency.
conda + pip ready.

Modern Python via conda/miniconda, JupyterLab, classic Notebook, plus Nginx for TLS termination. Add CUDA images on GPU plans.

One-click marketplace images on every plan
JupyterLab
Latest stable
Notebook
Classic interface
Python 3.12
Via conda or apt
conda
Env + package manager
pip
Standard PyPI installer
JupyterHub
Multi-user mode
CUDA
GPU acceleration (GPU plans)
Nginx
TLS reverse proxy

Use cases

Real Jupyter workloads
that need a real VPS.

Data analysis on remote data

When your dataset is 50 GB and your laptop has 16 GB RAM. Spin up a 32 GB VPS, scp the data, work over JupyterLab. Tear down when done.

ML model training

GPU plans with CUDA pre-installed give you A100 or RTX 5090 firepower for training. JupyterLab + PyTorch in your browser, no Colab limits.

Long-running computations

Some experiments run for days. Fire-and-forget on a Cloudzy VPS - no laptop sleep, no broken kernel, no cap on session length.

Team collaboration via JupyterHub

JupyterHub on the 12 GB plan lets your team share kernels and data without emailing notebooks around. Each user gets isolated workdirs.

Teaching / workshops

Spin up matching Jupyter environments for a workshop. Hourly billing means you only pay for the actual session hours.

API + notebook combo

Develop a model in Jupyter, expose the trained version as a Flask/FastAPI endpoint on the same VPS. One box, two endpoints.

60s
Provisioning
40 Gbps
Uplink
NVMe-only
Storage
12
منطقه‌ها
99.95%
آپتایم SLA
14 days
بازگشت وجه

شبکه جهانی

12 regions. Four continents.
One click away.

Drop your Jupyter VPS as close to your users as physics allows.

See all 12 regions
us-utah-1us-dal-1us-lax-1us-nyc-1us-mia-1eu-ams-1eu-lon-1eu-fra-1eu-zrh-1me-dxb-1ap-sgp-1ap-tyo-1

قیمت‌گذاری

فقط برای آنچه مصرف می‌کنید بپردازید. That's it.

Hourly, monthly, or yearly. Currently ۵۰٪ تخفیف all plans.

1 GB DDR5

Light · Personal use

$3.48 /ماه
$6.95/mo ۵۰%-
Deploy now
۱۴ روز ضمانت بازگشت وجه
  • 1 vCPU @ EPYC
  • 25 GB NVMe
  • 1 TB · 40 Gbps
  • Dedicated IPv4 + IPv6
  • Root SSH · KVM
2 GB DDR5

Production · Small team

$7.475 /ماه
$14.95/mo ۵۰%-
Deploy now
۱۴ روز ضمانت بازگشت وجه
  • 1 vCPU @ EPYC
  • 60 GB NVMe
  • 3 TB · 40 Gbps
  • Dedicated IPv4 + IPv6
  • Root SSH · KVM
محبوب‌ترین
4 GB DDR5

Production · Multi-site

$14.47 /ماه
$28.95/mo ۵۰%-
Deploy now
۱۴ روز ضمانت بازگشت وجه
  • 2 vCPU @ EPYC
  • 120 GB NVMe
  • 5 TB · 40 Gbps
  • Dedicated IPv4 + IPv6
  • Root SSH · KVM
8 GB DDR5

Heavy workloads

$26.475 /ماه
$52.95/mo ۵۰%-
Deploy now
۱۴ روز ضمانت بازگشت وجه
  • 4 vCPU @ EPYC
  • 240 GB NVMe
  • 7 TB · 40 Gbps
  • Dedicated IPv4 + IPv6
  • Root SSH · KVM
Need more? See all 10 plans (up to 64 GB)

FAQ. Jupyter VPS

Common questions, straight answers.

What is a Jupyter Notebook VPS?
A Jupyter Notebook VPS is a virtual private server where you run JupyterLab or classic Jupyter Notebook over a TLS-secured URL. You access it from any browser, you keep root SSH for kernel management, and you tear it down when the project ends.

How do I install Jupyter?
Install miniconda, then `conda install -c conda-forge jupyterlab`. Configure a Nginx reverse proxy with Certbot TLS, set a notebook password, point your DNS at the VPS. End-to-end takes about 20 minutes.

Can I run JupyterHub for a team?
Yes. JupyterHub installs cleanly on the 8-12 GB plans. Each user gets a single-user Jupyter server. Pair with PAM or OAuth for authentication. Most teams use the standard tljh (The Littlest JupyterHub) installer.

What about GPU acceleration for ML?
Pick a Premium GPU plan instead of regular CPU. Cloudzy GPU plans include Nvidia A100, RTX 5090, and RTX 4090 with CUDA pre-installed. Run PyTorch, TensorFlow, or JAX directly in your notebooks.

Is there a way to access notebooks without TLS?
Yes - SSH tunnel: `ssh -L 8888:localhost:8888 root@your-vps`, then open localhost:8888 in your browser. Skip the Nginx + TLS setup entirely if you only need solo access.

Will my notebooks survive a reboot?
Yes. Notebooks live on disk - reboots don't erase them. Run JupyterLab as a systemd service so it auto-starts. Snapshot the VPS for extra safety.

How much RAM should I pick?
Depends on dataset size. 2 GB for tutorials, 4-8 GB for typical pandas + scikit-learn work, 16-32 GB for large dataframes or deep learning on CPU. GPU plans for actual training.

Is there a money-back guarantee?
Yes - 14 days, full refund, no questions asked. Cancel from the panel anytime in the first two weeks.

آماده‌ایم وقتی شما آماده‌اید.
Jupyter VPS in 60 seconds.

Pick a region, click deploy. We'll have you working before lunch.

Deploy Jupyter VPS مقایسه همه پلان‌ها

بدون نیاز به کارت اعتباری · ضمانت بازگشت وجه ۱۴ روزه · لغو در هر زمان