Skip to content

OPTIMADE Python tools

Latest release Build status Activity
PyPI Version
PyPI - Python Version
Build Status
Heroku App Status
Commit Activity
Last Commit

The aim of OPTIMADE is to develop a common API, compliant with the JSON:API 1.0 specification. This is to enable interoperability among databases that contain calculated properties of existing and hypothetical materials.

This repository contains a library of tools for implementing and consuming OPTIMADE APIs using Python. Server implementations can make use of the supported MongoDB (v4) and Elasticsearch (v6) database backends, or plug in a custom backend implementation. The package also contains a server validator tool, which may be called from the shell (optimade-validator) or used as a GitHub Action from optimade-validator-action.

The release history and changelog can be found in the changelog.


This document, guides, and the full module API documentation can be found online at In particular, documentation of the OPTIMADE API response data models (implemented here with pydantic) can be found online under OPTIMADE Data Models.


Detailed installation instructions for different use cases (e.g., using the library or running a server) can be found in the installation documentation.

The latest stable version of this package can be obtained from PyPI pip install optimade. The latest development version of this package can be installed from the master branch of this repository git clone

Supported OPTIMADE versions

Each release of the optimade package from this repository only targets one version of the OPTIMADE specification, summarised in the table below.

OPTIMADE API version optimade version
v1.0.0 v0.12.9
v1.1.0 v0.16.0

Contributing and Getting Help

All development of this package (bug reports, suggestions, feedback and pull requests) occurs in the optimade-python-tools GitHub repository. Contribution guidelines and tips for getting help can be found in the contributing notes.

How to cite

If you use this package to access or serve OPTIMADE data, we kindly request that you consider citing the following: