Mars
Mars provides tensor-based tools for large-scale computation with NumPy, Pandas, and scikit-learn.
About Mars
Mars is an open source Python framework for large-scale data computation. It provides NumPy-style tensors, pandas-style DataFrames, and scikit-learn-style machine learning tools, and it can run Python functions in parallel. The project describes it as a unified framework that scales NumPy, pandas, scikit-learn and other libraries.
Mars can be installed with pip (the package is named pymars). It can run locally, where the README says it can use multiple cores even on a laptop, or connect to an existing Mars cluster through its web address. The README says it could be faster in a distributed setting, but it does not give performance guarantees.
Key features
- NumPy-like tensor interface through mars.tensor
- pandas-like DataFrame interface through mars.dataframe
- scikit-learn-style machine learning tools through mars.learn
- Integrations with TensorFlow, PyTorch, XGBoost, LightGBM, Joblib and Statsmodels
- mars.remote for executing Python functions in parallel
- Local session startup or connection to an existing Mars cluster
Best for: Python developers who already use NumPy, pandas, or scikit-learn and want to run similar code on larger datasets.
Good for
- Interfaces are modeled on NumPy, pandas, and scikit-learn, so existing code patterns carry over
- Can use multiple CPU cores on a single machine, according to the README
- Open source
Mars at a glance
- Website
- github.com/mars-project/mars
- Category
- Development › Developer Tools
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
Guides for Developer Tools
Mars alternatives
Tools and sites you could use instead, ranked by how closely they match.
Automatic machine learning, hyperparameter tuning, and pipeline abstractions are provided by the Neuraxle software library.
Open source
24 alternatives
Python developers can use Zipline’s event-driven library to backtest algorithmic trading strategies.
Open source
24 alternatives
torchaudio provides data manipulation and transformation tools for audio signal processing with PyTorch.
Open source
24 alternatives
Vaex provides out-of-core DataFrames for Python to explore and visualize large tabular datasets, with support for machine learning workflows.
Wind data sources can be accessed from Python with windML, a framework built on NumPy, SciPy, scikit-learn, and Matplotlib.
Browse large JSONL, Parquet, and CSV datasets in VS Code with the Bread Dataset Viewer extension.
Open source
24 alternativesPopular in Developer Tools
Well-known listings from the same topic.
AppDevToolsappdevtools.comCommand Not Foundcommand-not-found.comTestFlighttestflight.apple.com/join/A8GwCnq8Sourcegraphabout.sourcegraph.comProject Wallaceprojectwallace.comCarbon Calculatorwebsitecarbon.com
Mars FAQ
How do I install Mars?
The README gives the command pip install pymars. Developers who want to contribute can clone the repository and run pip install -e ".[dev]".
How do I start a local Mars session?
Import mars and call mars.new_session(), which starts a runtime locally.
Can Mars connect to a cluster that is already running?
Yes. The README shows mars.new_session() being called with the web address of an initialized Mars cluster.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.