• Chainer: Powerful, Flexible, and Intuitive Framework for Neural Networks

• pytorch-pfn-extras: Supplementary components to accelerate research and development in PyTorch

• ffcv: Fast Forward Computer Vision - A drop-in data loading system that dramatically increases data throughput in model training

• Microsoft DeepSpeed: Deep learning optimization library that makes distributed training easy, efficient, and effective

• vLLM: Easy, fast, and cheap LLM serving for everyone

• spaCy: Industrial-strength Natural Language Processing (NLP) with Python and Cython

• Thinc: Machine Learning library for NLP in Python

• MONAI: AI Toolkit for Healthcare Imaging

• py4DSTEM: Set of python tools for processing and analysis of four-dimensional scanning transmission electron microscopy (4D-STEM) data

• pomegranate: Fast, flexible and easy to use probabilistic modelling in Python

• NVIDIA RAPIDS: Open GPU Data Science

  • cuDF: GPU DataFrame library for loading, joining, aggregating, filtering, and otherwise manipulating data
  • cuML: Suite of libraries that implement machine learning algorithms and mathematical primitives functions
  • cuSignal: GPU-accelerated signal processing library
  • cuCIM: Toolkit providing GPU-accelerated I/O, computer vision & image processing primitives for N-Dimensional images with a focus on biomedical imaging
  • cuxfilter: Framework to connect web visualizations to GPU accelerated crossfiltering
  • kvikio: C++ and Python bindings to cuFile

• NVTabular: Feature engineering and preprocessing library for tabular data

• XGBoost: Scalable, Portable and Distributed Gradient Boosting Library

• einops: Flexible and powerful tensor operations for readable and reliable code

• PyLops: A Linear-Operator Library for Python

• SigPy: Python package for signal processing, with emphasis on iterative methods

• xarray: N-D labeled arrays and datasets

• xarray-spatial: Raster-based Spatial Analytics for Python

• scikit-learn: Machine learning library built on top of SciPy (throughArray API support)

• TensorLy: Tensor Learning in Python

• Zarr: An implementation of chunked, compressed, N-dimensional arrays for Python.

• Dask: Scalable analytics in Python

• Ray: An open source framework that provides a simple, universal API for building distributed applications

• NVIDIA cuQuantum: SDK of optimized libraries and tools for accelerating quantum computing workflows

• Optimized Einsum (opt_einsum): A tensor contraction order optimizer

• qutip-cupy: CuPy backend plug-in for QuTiP (Quantum Toolbox in Python)

• qibo: A framework for quantum computing with hardware acceleration

• quimb: Library for quantum information and many-body calculations

• vaex: High Performance Data Visualization

• PyQtGraph: A pure-Python graphics library for PyQt5/PyQt6/PySide2/PySide6

• napari-cupy-image-processing: napari Plug-in for GPU-accelerated image processing

• turboSETI: Python based SETI search algorithm

• 3d-ken-burns: 3D Ken Burns Effect from a Single Image using PyTorch

• pytorch-qrnn: Quasi-Recurrent Neural Network (QRNN) for PyTorch

• Kubernetes: Production-Grade Container Scheduling and Management (uses CuPy as a tool forend-to-end testing of CUDA support)

Seerepositories tagged with cupy andDependents to explore more projects using/supporting CuPy.

• NLCPy byNEC Corporation: CuPy for SX-Aurora TSUBASA
• ClPy byFixstars: CuPy for OpenCL (inactive)
• cupy-ibmopt byIBM: CuPy optimized for POWER architecture (inactive)