---
id: opt-einsum
version: "3.4.0"
license: unclear
license_treatment: permissive
maintenance: dormant
---

# opt-einsum — Path optimization of einsum functions.

License: permissive · Maintenance: dormant · Popularity: top 1,000 on PyPI

## Install

pip install opt-einsum
uv add opt-einsum
poetry add opt-einsum

## Description

# Optimized Einsum

[![Tests](https://github.com/dgasmith/opt_einsum/actions/workflows/Tests.yml/badge.svg)](https://github.com/dgasmith/opt_einsum/actions/workflows/Tests.yml)
[![codecov](https://codecov.io/gh/dgasmith/opt_einsum/branch/master/graph/badge.svg)](https://codecov.io/gh/dgasmith/opt_einsum)
[![Anaconda-Server Badge](https://anaconda.org/conda-forge/opt_einsum/badges/version.svg)](https://anaconda.org/conda-forge/opt_einsum)
[![PyPI](https://img.shields.io/pypi/v/opt_einsum.svg)](https://pypi.org/project/opt-einsum/#description)
[![PyPIStats](https://img.shields.io/pypi/dm/opt_einsum)](https://pypistats.org/packages/opt-einsum)
[![Documentation Status](https://github.com/dgasmith/opt_einsum/actions/workflows/Docs.yaml/badge.svg)](https://dgasmith.github.io/opt_einsum/)
[![DOI](https://joss.theoj.org/papers/10.21105/joss.00753/status.svg)](https://doi.org/10.21105/joss.00753)

## Optimized Einsum: A tensor contraction order optimizer

Optimized einsum can significantly reduce the overall execution time of einsum-like expressions...

## AI interpretation — verify before relying

Optimizes the execution order of einsum tensor contractions to dramatically reduce computation time, acting as a drop-in replacement for einsum functions across multiple tensor libraries.

Verdict: A well-established, dependency-free tensor optimization library in Production/Stable status with no known vulnerabilities and strong historical adoption (top 1000 PyPI). Its dormancy (686 days since release) is a minor concern for active development, but the mature codebase and broad backend support make it reliable for einsum optimization tasks.

[View on SkillFed](https://skillfed.io/packages/opt-einsum) · [View on PyPI](https://pypi.org/project/opt-einsum/)