awesome-machine-learning

A language by language directory of machine learning frameworks and libraries, in the spirit of awesome-php, whose owner says the repository is not maintained and has grown tired of PRs from LLMs.

The repository is a curated list of machine learning frameworks, libraries, and software, organized by programming language and inspired by awesome-php. It functions as a starting point for finding machine learning software across different languages. Entries are short descriptions with links to each project, so the list acts as a directory rather than hosting any code itself. Python is listed as the primary language, and there is no standard license identifier.

A note dated April 2026 explains the situation plainly. The owner writes that too many PRs are being generated by LLMs and that it is no longer fun or manageable. Contributors who want to add to the repo are asked to email the owner with a link to their PR to prove they are human before it will be merged. A second note states the owner says the library is not maintained and has not had commits for a long time, around two to three years.

The C section covers general purpose machine learning libraries. Darknet is described as an open source neural network framework written in C and CUDA that is fast, easy to install, and supports CPU and GPU computation. cONNXr is an ONNX runtime written in pure C with zero dependencies, aimed at small embedded devices. Computer vision entries include CCV and VLFeat, and some entries carry a deprecated tag.

The C++ section is longer. It lists DLib, OpenCV, Openpose, Caffe, CatBoost, and CNTK among the deep learning and computer vision libraries. It continues with EBLearn, VIGRA, the BLLIP parser, and colibri-core. Each entry keeps the same short format: a name, a one line description, and a link. Several are marked deprecated.

Organization follows language. The list starts with APL, moves to C, then C++, and continues with other languages further down. Descriptions vary in length but the format stays consistent across sections. The README frames the whole thing as a directory of links to each framework rather than a place that hosts code, in the same spirit as the awesome-php list that inspired it.

The list still works as a directory of links to frameworks organized by language. Anyone who wants to contribute now has to email the owner first to prove they are human.

• Source 1 · Project repository
• Source 2 · README