Jetbrains Datalore

What is included in Jetbrains Datalore?

Jupyter Notebooks – Edit and run native .ipynb files.
Multi-Language Kernels – Python, SQL, R, Scala, and Kotlin.
SQL Cells – Query connected databases directly inside notebooks.
Interactive Reports – Publish notebooks as shareable interactive reports.
Core Capacity – Five notebook languages with real-time collaboration.
Important – Browser-based platform; no offline desktop app.

What are the main benefits of Jetbrains Datalore?

Datalore is JetBrains' collaborative, Jupyter-compatible data science platform for writing, running, and sharing notebooks in the cloud or self-hosted on-premises. It combines PyCharm-grade coding assistance with managed CPU and GPU machines, native database queries, and one-click report publishing.

Real-Time Collaboration – Multiple users edit one notebook simultaneously.
Managed Compute – Run code on cloud CPUs and GPUs.
Datalore AI – Generates and fixes Python, SQL, R code.
Database Connectivity – Native SQL cells query connected sources.
PyCharm Code Insight – Completion, refactorings, and quick-fixes while coding.
Self-Hosting Option – On-Premises keeps data on your infrastructure.

What does Jetbrains Datalore do?

Datalore is a cloud-based and self-hostable platform for working with Jupyter notebooks in Python, SQL, R, Scala, and Kotlin. It runs your code on JetBrains-managed CPU and GPU machines, so heavy data processing or model training does not depend on your local laptop. Native SQL cells let you query a connected database and pass the result straight into a pandas DataFrame in the same notebook. You can then publish the notebook as an interactive report or data app and share it by link, with code cells hidden from viewers.

Who is Jetbrains Datalore best suited for?

Datalore fits data teams that want a shared notebook environment instead of each analyst maintaining a separate local Jupyter setup. Because workspaces hold notebooks, data, and environments together, it removes the version-mismatch problems that occur when colleagues install packages individually. The integrated package manager pins dependencies so a notebook reopens with the same environment, which matters for reproducible analysis hand-offs. Teams with strict data-residency requirements can run the identical platform through the On-Premises edition on their own servers.

How do the Datalore Cloud Free, Cloud, and On-Premises plans differ?

The three editions share the same notebook editor but differ in compute, AI, and sharing rights. Cloud Free runs up to two notebooks in parallel, shares notebooks and workspaces with view-only access, and excludes Datalore AI, version history, and GPU machines. The paid Cloud plan adds Datalore AI, Reactive mode, unlimited parallel and scheduled notebooks, version history, and access to powerful CPU and GPU machines. On-Premises (previously called Enterprise) unlocks every feature and adds an administration interface plus SSO, letting you host the whole platform on your own infrastructure.

What is missing from the Datalore Cloud Free plan?

On Cloud Free, notebook and workspace sharing is view-only, so collaborators cannot edit your notebooks until you move to the paid Cloud plan. The free tier also excludes Datalore AI, version history, Reactive mode, scheduled runs, and GPU machines, and limits you to two notebooks running in parallel. No edition ships an offline desktop application — Datalore always runs in the browser, even when self-hosted through On-Premises. If your team needs editable collaboration or AI-assisted code generation, Cloud Free will not be sufficient.

Which languages, databases, and file formats does Datalore support?

Datalore notebooks support Python, SQL, R, Scala, and Kotlin, and they read and write standard .ipynb files for portability with Jupyter. Native SQL cells connect to databases such as PostgreSQL, MySQL, Microsoft SQL Server, Snowflake, BigQuery, and Redshift, with query results flowing directly into pandas. You can mount cloud storage like Amazon S3 and attach Git repositories to a notebook to install custom pip-compatible packages. Visualization works with matplotlib, plotly, altair, seaborn, and lets-plot, alongside built-in line, bar, area, and correlation charts.

What should users check before choosing Jetbrains Datalore?

Confirm whether you need editable collaboration, Datalore AI, or GPU compute, because all three require the paid Cloud plan rather than Cloud Free. If your organization must keep data inside its own network, plan for the On-Premises edition, which also enables SSO through SAML, Okta, or Azure AD. Verify that the databases your team uses appear in Datalore's native connector list, since each connection is configured per workspace. Finally, remember the platform is browser-based, so it does not replace a local desktop IDE such as PyCharm.

Frequently asked questions about Jetbrains Datalore

Is Jetbrains Datalore compatible with existing Jupyter notebooks?

Yes. Datalore imports and exports standard .ipynb files and offers a Jupyter computation mode, so existing notebooks open without conversion. On the paid Cloud plan you can also switch a notebook to Reactive mode for automatic dependency-based cell re-execution.

Does Datalore keep data private and compliant?

Datalore's cloud service is GDPR-compliant and SOC 2 Type II certified. For tighter control, the On-Premises edition hosts notebooks and data entirely on your own infrastructure and supports SSO via SAML, Okta, and Azure AD.

Can Datalore connect to Git repositories?

Yes. You can attach a Git branch to a notebook to install a custom pip-compatible package, and workspace contents can be synced to Git repositories for versioned backups and collaboration with engineering teams.