PyCaret
PyCaretLow-code machine-learning library that wraps scikit-learn, XGBoost, LightGBM, CatBoost and other libraries behind a few calls. After a setup step it trains and ranks every applicable model with one function, then tunes, blends, evaluates and deploys the chosen pipeline, for classification, regression, clustering, anomaly detection and time-series forecasting. The 4.0 line in development adds a web control plane and moves the engine to the Functional Source License.
Openness
5 medium confidence- license
- MIT(OSI
- source
- public(github.com/pycaret/pycaret (tag 3.3.2))
- core features withheld
- no — no paid tier in the 3.x library
- upcoming-terms
- FSL-1.1-MIT(main branch and the 4.0 alpha pre-releases
The release people install today, PyCaret 3.3.2, is MIT-licensed and builds from the public repository. The maintainers have since relicensed the engine on the main branch to the Functional Source License (FSL-1.1-MIT), which forbids competing commercial use for two years after each release, starting with PyCaret 4.0. Only alpha builds of 4.0 have shipped, so the MIT release governs until 4.0 is released.
- https://pypi.org/pypi/pycaret/4.0.0a8/json recorded 2026-09-27
PyPI pre-release 4.0.0a8, uploaded 2026-05-08: license_expression "LicenseRef-FSL-1.1-MIT".
- https://pypi.org/pypi/pycaret/json recorded 2026-09-27
PyPI: latest stable version 3.3.2, uploaded 2024-04-28, license MIT, classifier "License :: OSI Approved :: MIT License"; release list continues only with pre-releases 4.0.0a3 to 4.0.0a8.
- https://raw.githubusercontent.com/pycaret/pycaret/3.3.2/LICENSE recorded 2026-09-27
LICENSE at tag 3.3.2 is the "MIT License", "Copyright (c) 2024 PyCaret".
- https://raw.githubusercontent.com/pycaret/pycaret/3.3.2/README.md recorded 2026-09-27
README at 3.3.2: "PyCaret is an open-source, low-code machine learning library in Python that automates machine learning workflows."; no paid tier or commercial edition mentioned.
- https://raw.githubusercontent.com/pycaret/pycaret/main/docs/revamp/DECISIONS.md recorded 2026-09-27
DECISIONS.md, 2026-04-27: "Engine moves from MIT to FSL-1.1-MIT ... starting with PyCaret 4.0. The 3.x line on PyPI (pycaret <= 3.4.0) stays MIT - the relicense applies prospectively to 4.0+ only."
- https://raw.githubusercontent.com/pycaret/pycaret/main/LICENSE recorded 2026-09-27
LICENSE on main: "Functional Source License, Version 1.1, MIT Future License", abbreviation "FSL-1.1-MIT", "Copyright 2026 Moez Ali and PyCaret contributors"; a Permitted Purpose is "any purpose other than a Competing Use".
- https://ungh.cc/repos/pycaret/pycaret/files/main recorded 2026-09-27
Repository tree of 631 paths with no ee, enterprise, commercial or pro directory.
Adoption
3 high confidenceMeasured on monthly PyPI downloads of the pycaret package, almost all of them the 3.x line.
- https://pypistats.org/api/packages/pycaret/recent recorded 2026-09-27
last_month 225,282 downloads of pycaret
Capability
5 medium confidencePyCaret's compare_models call trains and ranks every applicable model on the user's data and returns the best, so choosing among learners is done for the user. It sits with AutoGluon on that automation, though the user still drives each step with a function call.
- https://raw.githubusercontent.com/pycaret/pycaret/3.3.2/README.md recorded 2026-09-27
README at 3.3.2 quickstart: "s = setup(data, target = 'Purchase', session_id = 123)", "# model training and selection", "best = compare_models()", then evaluate_model, predict_model and save_model; "PyCaret is essentially a Python wrapper around several machine learning libraries and frameworks such as scikit-learn, XGBoost, LightGBM, CatBoost, Optuna, Hyperopt, Ray".
Verified 2026-09-27