GPU-powered numerical computing library
CuPy combines the numerical computing library NumPy and SciPy interfaces with graphics processing unit (GPU) acceleration. This Python-based library accelerates complex mathematical operations using CUDA cores.
Updates
- August 20, 2026: Stars 12,227 → 12,259, latest release v14.2.0 (August 20, 2026).
- August 2, 2026: Stars 11,639 → 12,227, latest release v14.1.1 (June 1, 2026).
What you get
- Compatible with NumPy and SciPy
- NVIDIA CUDA and AMD ROCm support
- Access to low-level CUDA features
Running it
docker run --gpus all -it cupy/cupyIf you don't write code
I want to take advantage of GPU acceleration in my Python code instead of NumPy or SciPy. How can I run my existing NumPy codes on the GPU using CuPy and integrate low-level CUDA features like RawKernels into my project?
Repository: https://github.com/cupy/cupy. Follow the official instructions in the README of this repository for setup and execution.
Before starting, check which OS we are on and if the necessary tools (e.g., Python, Node.js, Git, Docker) are installed. Install any missing ones step-by-step according to this system.
Create a new, empty folder to work in. Do not touch any files outside this folder and do not delete anything. Tell me what you are going to do in one sentence before running each command.
Tell me what I should see on the screen at the end of each step. If an error occurs, I will paste the error message to you exactly as it is, and we will fix it together.
Related dictionary terms
Links
TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-06-29: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. The introductory text was prepared with AI from the repository's own documentation, and this page was machine-translated from the Turkish original · the Turkish version prevails.