Training and inference frameworks for developers, requiring your own data and compute.
Machine learning frameworks rankings
Machine learning frameworks, ordered by GitHub stars · 10 projects
Sort
1
pytorchpytorch 104.9kPyTorch is a Python deep learning framework providing GPU-accelerated tensor computation and tape-based autograd for training neural networks in research and production.Python 2026.09.302
pytorch-lightningLightning-AI 31.4kPyTorch Lightning is a deep learning framework (Python) that automates engineering code such as backpropagation, mixed precision, multi-GPU, and distributed training, so the same model code scales from 1 to 10,000+ GPUs with zero changes.Python 2026.09.103
xgboostdmlc 28.9kXGBoost is an optimized distributed gradient boosting library (C++ core with Python, R, Java, and other bindings) that solves classification and regression on tabular data with efficient parallel tree boosting, scaling to billions of examples.C++ 2026.09.154
onnxruntimemicrosoft 22.2kMicrosoft's open-source cross-platform machine learning inference and training accelerator, supporting PyTorch, TensorFlow/Keras, and classical ML libraries like scikit-learn and LightGBM.C++ 2026.10.265
LightGBMlightgbm-org 18.8kA high-performance, distributed gradient boosting (GBDT) framework based on decision tree algorithms, used for ranking, classification, and other machine learning tasks.C++ 2026.07.186
mujocogoogle-deepmind 15.8kA general-purpose physics engine maintained by Google DeepMind for fast and accurate simulation in robotics, biomechanics, graphics, and machine learning.C++ 2026.10.227
Cirqquantumlib 5.1kCirq is an open-source Python framework by Google for creating, editing, and running NISQ quantum circuits on simulators or real quantum hardware.Python 2026.06.308
strawberrystrawberry-graphql 4.7kStrawberry is a Python GraphQL library based on type annotations, defining schemas in a dataclasses style, with integrations for Django, FastAPI, Starlette, and more.Python 2026.10.269
implicitbenfred 3.8kimplicit is a fast collaborative filtering library for implicit feedback datasets (Python), providing ALS, BPR, logistic matrix factorization, and item-nearest-neighbour models with multi-threaded CPU training and optional GPU acceleration.Python 2026.05.0810
determineddetermined-ai 3.2kAn open-source machine learning platform that simplifies distributed training, hyperparameter tuning, experiment tracking, and resource management, compatible with PyTorch and TensorFlow.Go 2025.03.19