Jupyter hosting
A persistent notebook server that does not sleep when your laptop does.
Jupyter hosting at AIHostBox means a dedicated, single-tenant Jupyter instance on our infrastructure: we install it, pin it to a reviewed release, put it behind TLS on your domain, back it up nightly and keep it patched. Plans start at $5.99 per month and the admin account, the data and any API keys are yours.
Jupyter is the notebook environment maintained by Project Jupyter, a community-governed open-source project that also maintains Jupyter Server and JupyterHub, and JupyterLab is its current interface: notebooks, a terminal, a file browser and text editor in one browser tab. Kernels run Python by default and other languages through additional kernels. Unlike hosted notebook services with idle timeouts, a Jupyter server on your own instance runs until you stop it, with no session limit and no per-hour compute meter.
The server model matters for long work. A kernel is a process on the machine, so on a laptop it dies with the lid or the battery. Hosted, a training run, scrape or batch of API calls keeps going after you close the tab, and the outputs are still in the notebook when you reconnect from another computer. The data you work on stays on the instance instead of being copied onto every analyst’s machine.
Jupyter hosting plans
Every plan is a dedicated single-tenant instance with NVMe storage, nightly off-box backups, TLS and managed patching. Quarterly, semi-annual and annual terms are discounted at checkout. Model tokens are not included — you bring your own key.
How we run Jupyter
A single JupyterLab server from the official Jupyter Docker Stacks images (quay.io/jupyter) on port 8888, with /home/jovyan on a persistent volume.
What we configure for you
- Password authentication through Jupyter Server’s PasswordIdentityProvider instead of a URL token
- DOCKER_STACKS_JUPYTER_CMD=lab so JupyterLab opens by default
- Image variant matched to your work, for example scipy-notebook for pandas and scikit-learn
- NB_UID and file ownership aligned with SFTP so uploaded files are writable in notebooks
- RESTARTABLE=yes so the server comes back if Jupyter exits
Which plan fits
Starter (2 GB, 1 vCPU) suits notebooks that call APIs and handle modest dataframes. Plus (4 GB, 2 vCPU) is for pandas work on datasets of a few hundred megabytes and scikit-learn models. Pro (8 GB, 3 vCPU) handles larger in-memory datasets and parallel jobs; if a dataset does not fit in RAM, process it in chunks or with DuckDB rather than expecting a bigger plan to fix it.
The honest limits
There is no GPU, so deep-learning training is impractical beyond small experiments. One server means one user identity; multi-user separation needs JupyterHub.
What people run Jupyter for
Long jobs
Start an analysis and close the lid.
Shared environment
One reproducible setup instead of six local ones.
LLM prototyping
Call your model provider or a private Ollama plan from notebooks, next to the data.
Scheduled data pulls
Run a notebook-driven script from the terminal and read the results in the morning.
Licence and version policy
| Item | Position |
|---|---|
| Upstream licence | BSD-3-Clause |
| Sold as managed hosting | Yes |
| Tenancy | Single tenant — one customer per instance |
| Patching | Critical advisories within 24 hours, high severity within 72 |
Our full position on licences, including the applications we decline to sell, is on the licensing page.
Works well with Jupyter
JupyterHub
Notebook servers for a whole team or classroom.
from $18.99/moDetails →
Ollama
Your private LLM server — pull a model, get an API, keep every token on your own box.
from $16.99/moDetails →
MLflow
Experiment tracking and a model registry that outlive the notebook.
from $5.99/moDetails →
Qdrant
A fast vector database with real filtering, written in Rust.
from $5.99/moDetails →More ML platform we host
Jupyter hosting: your questions
How is the notebook server protected?
Do installed packages survive restarts and upgrades?
Can several people share one Jupyter server?
Can I use large language models from notebooks?
How do I get data in and out?
Run Jupyter without running a server
A dedicated Jupyter instance, deployed and maintained by us. Month to month, cancel whenever.