NVIDIA Developer Zone

GPU-Accelerated Libraries

Adding GPU-acceleration to your application can be as easy as simply calling a library function. Check out the extensive list of high performance GPU-accelerated libraries below. If you would like other libraries added to this list please  contact us.


NVIDIA cuFFT
NVIDIA CUDA Fast Fourier Transform Library (cuFFT) provides a simple interface for computing FFTs up to 10x faster, without having to develop your own custom GPU FFT implementation.

NVIDIA cuBLAS
NVIDIA CUDA BLAS Library (cuBLAS) is a GPU-accelerated version of the complete standard BLAS library that delivers 6x to 17x faster performance than the latest MKL BLAS.

CULA Tools
GPU-accelerated linear algebra library by EM Photonics, that utilizes CUDA to dramatically improve the computation speed of sophisticated mathematics.

MAGMA
A collection of next gen linear algebra routines. Designed for heterogeneous GPU-based architectures. Supports current LAPACK and BLAS standards.

IMSL Fortran Numerical Library
Developed by RogueWave, a comprehensive set of mathematical and statistical functions that offloads  work to GPUs.

NVIDIA cuSPARSE
NVIDIA CUDA Sparse (cuSPARSE) Matric library provides a collection of basic linear algebra subroutines used for sparse matrices that delivers over 8x performance boost.

NVIDIA CUSP
A GPU accelerated Open Source C++ library of generic parallel algorithms for sparse linear algebra and graph computations. Provides an easy to use high-level interface.

AccelerEyes ArrayFire
Comprehensive GPU function library, including functions for math, signal and  image processing, statistics, and more.  Interfaces for C, C++, Fortran, and Python.

NVIDIA cuRAND
The CUDA Random Number Generation library performs high quality GPU-accelerated random number generation (RNG) over 8x faster than typical CPU only code.

NVIDIA NPP
NVIDIA Performance Primitives is a GPU accelerated library with a very large collection of 1000's of image processing primitives and  signal processing primitives.

NVIDIA CUDA Math Library
An industry proven, highly accurate collection of standard mathematical functions, providing high performance on NVIDIA GPUs.

Thrust
A powerful, open source library of parallel algorithms and data structures. Perform GPU-accelerated sort, scan, transform, and reductions with just a few lines of code.

GPU AI – Path Finding
Libraries and samples applications demonstrating CUDA-accelerated path planning suitable for domains such as robotics, video games, synthetic environments and AI.

GPU AI for Board Games
GPU accelerated tree search using pruning and backtracking.  Primarily focused on zero-sum games, also applicable to other domains including economics, environmental, psychological behavior, and AI.

Looking for More?
Know of a high value CUDA accelerated library that should be listed here? - Great! Contact us we want to know more.

Looking For More?

 

Find More CUDA Accelerated Libraries and Research Papers

Google Scholar indexes scholarly literature across disciplines. Suggested searches to get you started:
supercomputing
seismic processing
fluid dynamics		 computational chemistry
molecular dynamics
computational finance