The CUDA Library Samples repository contains various examples that demonstrate the use of GPU-accelerated libraries in CUDA. These libraries enable high-performance computing in a wide range of applications, including math operations, image processing, signal processing, linear algebra, and compression. The samples included cover:
- Math and Image Processing Libraries
- cuBLAS (Basic Linear Algebra Subprograms)
- cuTENSOR (Tensor Linear Algebra)
- cuSPARSE (Sparse Matrix Operations)
- cuSOLVER (Dense and Sparse Solvers)
- cuFFT (Fast Fourier Transform)
- cuRAND (Random Number Generation)
- NPP (Image and Video Processing)
- nvJPEG (JPEG Encode/Decode)
- nvCOMP (Data Compression)
- and more...
The CUDA Library Samples are provided by NVIDIA Corporation as Open Source software, released under the Apache 2.0 License. These examples showcase how to leverage GPU-accelerated libraries for efficient computation across various fields.
For more information on the available libraries and their uses, visit GPU Accelerated Libraries.
Explore the examples of each CUDA library included in this repository:
- cuBLAS - GPU-accelerated basic linear algebra (BLAS) library
- cuBLASLt - Lightweight BLAS library
- cuBLASMp - Multi-process BLAS library
- cuBLASDx - Device-side BLAS extensions
- cuDSS - GPU-accelerated linear solvers
- cuFFT - Fast Fourier Transforms
- cuFFTMp - Multi-process FFT
- cuFFTDx - Device-side FFT extensions
- cuPQC - Post-Quantum Cryptography device library
- cuRAND - Random number generation
- cuSOLVER - Dense and sparse direct solvers
- cuSOLVERMp - Multi-process solvers
- cuSOLVERSp2cuDSS - Transition example from cuSOLVERSp/Rf to cuDSS
- cuSPARSE - BLAS for sparse matrices
- cuSPARSELt - Lightweight BLAS for sparse matrices
- cuTENSOR - Tensor linear algebra library
- cuTENSORMg - Multi-GPU tensor linear algebra
- NPP - GPU-accelerated image, video, and signal processing functions
- NPP+ - C++ extensions for NPP
- nvJPEG - High-performance JPEG encode/decode
- nvJPEG2000 - JPEG2000 encoding/decoding
- nvTIFF - TIFF encoding/decoding
- nvCOMP - Data compression and decompression
Each sample provides a practical use case for how to apply these libraries in real-world scenarios, showcasing the power and flexibility of CUDA for a wide variety of computational needs.
For more information and documentation on CUDA libraries, please visit:
We welcome contributions to CUDA Library Samples. To contribute to CUDA Library Samples and make pull requests, follow the guidelines outlined in the Contributing document.
The CUDA Library Samples are distributed under the Apache 2.0 License. For more details, refer to the LICENSE.md file.
The old code that was originally distributed under the 3-clause "New" BSD license is available at bsd3_main branch and is no longer maintained.
