Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
CuPy
open-in-new
<p>CuPy is an <a href="/facts/Open-source_software/HzTaB8E9">open source</a> library for GPU-accelerated computing with <a href="/facts/Python_/YbuGqofa">Python</a> programming language, providing support for multi-dimensional arrays, sparse matrices, and a variety of numerical algorithms implemented on top of them. CuPy shares the same API set as <a href="/facts/NumPy/T6FnhWWD">NumPy</a> and <a href="/facts/SciPy/bNcMQGc0">SciPy</a>, allowing it to be a drop-in replacement to run NumPy/SciPy code on GPU. CuPy supports <a href="/facts/Nvidia/8QfCgqbu">Nvidia</a> <a href="/facts/CUDA/fvPpolot">CUDA</a> GPU platform, and <a href="/facts/AMD/CB1G7QTb">AMD</a> <a href="/facts/ROCm/MX8uasuY">ROCm</a> GPU platform starting in v9.0. </p><p>CuPy has been initially developed as a backend of <a href="/facts/Chainer/WN7uuwcz">Chainer</a> deep learning framework, and later established as an independent project in 2017. </p><p>CuPy is a part of the NumPy ecosystem array libraries and is widely adopted to utilize GPU with Python, especially in high-performance computing environments such as <a href="/facts/Summit_(supercomputer)/BmKWTpEk">Summit</a>, <a href="/facts/Perlmutter_(supercomputer)/g2nRPDQj">Perlmutter</a>, <a href="/facts/Brutus_cluster/eF0f6ySr">EULER</a>, and <a href="/facts/AI_Bridging_Cloud_Infrastructure/wMxM8gTh">ABCI</a>. </p><p>CuPy is a NumFOCUS sponsored project. </p>