cupy-cuda13x 14.0.1


pip install cupy-cuda13x

  Latest version

Released: Feb 20, 2026
Navigation
Meta
Author: Seiya Tokui
Maintainer: CuPy Developers
Requires Python: >=3.10
Classifiers

Development Status
  • 5 - Production/Stable

Intended Audience
  • Science/Research
  • Developers

Programming Language
  • Python
  • Python :: 3
  • Python :: 3.10
  • Python :: 3.11
  • Python :: 3.12
  • Python :: 3.13
  • Python :: 3.14
  • Python :: 3 :: Only
  • Cython

Topic
  • Software Development
  • Scientific/Engineering

Operating System
  • POSIX
  • Microsoft :: Windows
https://raw.githubusercontent.com/cupy/cupy/main/docs/image/cupy_logo_1000px.png

CuPy : NumPy & SciPy for GPU

CuPy is a NumPy/SciPy-compatible array library for GPU-accelerated computing with Python.

This is a CuPy wheel (precompiled binary) package for CUDA 13.x. You need to install CUDA Toolkit 13.x locally to use these packages. Alternatively, you can install this package together with all needed CUDA components from PyPI by passing the [ctk] tag:

$ pip install cupy-cuda13x[ctk]

If you have another version of CUDA, or want to build from source, refer to the Installation Guide for instructions.

14.0.1 Feb 20, 2026
14.0.0 Feb 17, 2026
13.6.0 Aug 18, 2025

Wheel compatibility matrix

Platform CPython 3.10 CPython 3.11 CPython 3.12 CPython 3.13 CPython 3.14
manylinux2014_aarch64
manylinux2014_x86_64
win_amd64

Files in release

Extras:
Dependencies:
numpy (<2.6,>=2.0)
cuda-pathfinder (==1.*,>=1.3.3)