autodiff ☆

« Back to VersTracker

Description:
Automatic differentiation made easier for C++

Type: Formula | Latest Version: 1.1.2@1 | Tracked Since: Dec 17, 2025

Links: Homepage | formulae.brew.sh

Category: Developer tools

Tags: c++ automatic-differentiation math scientific-computing machine-learning

Install: brew install autodiff

About:
autodiff is a C++ library that provides fast and accurate differentiation capabilities, essential for optimization and machine learning tasks. It leverages modern C++ features to compute derivatives of complex functions using forward, reverse, and higher-order modes. The library integrates seamlessly with Eigen for linear algebra operations, making it a powerful tool for scientific computing.

Key Features:

Supports forward, reverse, and higher-order automatic differentiation
Seamless integration with Eigen matrix library
High-performance computations with minimal overhead
Clean, modern C++11/14/17 API

Use Cases:

Optimizing machine learning model parameters
Solving systems of nonlinear equations in scientific computing
Robotics and control systems engineering
Financial modeling and risk analysis

Alternatives:

CppAD – CppAD is another mature C++ AD library, while autodiff offers a more modern API and better Eigen integration
Stan Math – Stan Math is focused on Bayesian statistics, whereas autodiff is a more general-purpose differentiation library

License: MIT

Dependencies: eigen@3, pybind11

Bottles available for: arm64_tahoe, arm64_sequoia, arm64_sonoma, sonoma, arm64_linux, x86_64_linux

Version History

Detected	Version	Rev	Change	Commit
Sep 13, 2025 1:02pm		0	VERSION_BUMP	409e2560