#deep-learning

Graph Neural Network Library for PyTorch

Streamlining reinforcement learning with RLOps. State-of-the-art RL algorithms and tools, with 10x faster training through evolutionary hyperparameter optimization.

CUDA Templates and Python DSLs for High-Performance Linear Algebra

Fast and accurate AI powered file content types detection

Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages

An Open Source Machine Learning Framework for Everyone

⚡️SwanLab - an open-source, modern-design AI training tracking and visualization tool. Supports Cloud / Self-hosted use. Integrated with PyTorch / Transformers / verl / LLaMA Factory / ms-swift / Ultralytics / MMEngine / Keras etc.

Relax! Flux is the ML library that doesn't make you tensor

Brevitas: neural network quantization in PyTorch

A CLI for Kubeflow.

TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and support for E2E production ML pipelines when you're ready.

High-performance automatic differentiation of LLVM and MLIR.

The fastai deep learning library

Collection of eclectic utils for python.

Image annotation with Python. Supports polygon, rectangle, circle, line, point, and AI-assisted annotation.

Image annotation with Python. Supports polygon, rectangle, circle, line, point, and AI-assisted annotation.

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

A Python library for anomaly detection across tabular, time series, graph, text, image, and audio data. 60+ detectors, benchmark-backed ADEngine orchestration, and an agentic workflow for AI agents.

SDK for running DeepLabCut on a live video stream

Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)

An open source implementation of CLIP.

Neural Network primitives with multiple backends

Fast inference engine for Transformer models

SKaiNET makes local AI practical for developers: simple to build with, multiplatform by design, and optimized for native performance without compromises.

Statistical Machine Intelligence & Learning Engine

Visualizer for neural network, deep learning and machine learning models

Type annotations and runtime checking for shape and dtype of JAX/NumPy/PyTorch/etc. arrays. https://docs.kidger.site/jaxtyping/

NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.

Tools for computational pathology

Faster structured generation