Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

AIX360

aix360.res.ibm.com

  • Free plan
  • Windows
  • macOS
  • Linux
  • API
  • Self-hosted

AIX360

Free

Free plan

What the maker publishes

Algorithm examples
Listed methods include LIME, SHAP, ProtoDash, and time-series explanation approaches. source
Algorithms and metrics
Its Python package includes algorithms covering different explanation dimensions and proxy explainability metrics. source
Audience and support
The toolkit provides data-scientist-oriented tutorials and directs contributors to a Slack community. source
Data types
The toolkit supports tabular, text, image, and time-series data. source
Dependencies
Optional algorithm-specific dependencies can require conflicting Python versions, so not every algorithm combination can share one environment. source
Extensibility
The library was developed with extensibility in mind and invites contributions of algorithms, metrics, and use cases. source
Guidance
The project provides guidance material and a taxonomy tree to help select explanation algorithms. source
Install
AIX360 can be installed from PyPI using pip install aix360. source
Platforms
The toolkit is tested on Windows, macOS, and Linux. source
Project status
The project describes the library as still in development. source
Purpose
AIX360 is an open-source Python toolkit for interpreting and explaining datasets and machine-learning models. source
Python versions
Python requirements vary by algorithm; the installation table lists versions including 3.6 and 3.10. source
Security and license
The repository links to a security policy and identifies an Apache-2.0 license. source
Tutorials
The repository includes example and tutorial Jupyter notebooks. source

Checked against the maker's pages on 14 Nov 2026.