NVIDIA GPU-accelerated libraries provide highly-optimized functions that perform 2x-10x faster than CPU-only alternatives. Using drop-in interfaces, you can replace CPU-only libraries such as MKL, IPP and FFTW with GPU-accelerated versions with almost no code changes. The libraries can optimally scale your application across multiple GPUs.
GPU-accelerated libraries for linear algebra, signal processing, image and video processing lay the foundation for compute-intensive applications in areas such as molecular dynamics, computational chemistry, medical imaging and seismic exploration. For deep learning, NVIDIA provides specialized libraries that are integrated with all the leading deep learning frameworks. NVIDIA libraries use optimized precision to maximize performance for applications such as computer vision, speech processing and natural language processing. Today, NVIDIA’s libraries are running from resource constrained IoT devices, to self driving cars, to the largest supercomputers on the planet.
With NVIDIA’s libraries, you get highly efficient implementations of algorithms that are regularly extended and optimized. Whether you are building a new application or trying to speed up an existing application, NVIDIA’s libraries provide the easiest way to get started with GPUs. You can download NVIDIA libraries as part of the CUDA Toolkit.Download Now CUDA 9.1: What's New...
NVIDIA libraries perform faster than alternatives through new algorithms and optimizations. The latest release, CUDA 8, is over 2x faster than its predecessor. As a user of the library, your application automatically benefits from such regular platform improvements without any changes necessary to your application code.
NVIDIA’s libraries help accelerate existing applications with minimal code changes. You can use them with languages that you already know, including C, C++, Fortran, Python and MATLAB. The libraries work across all NVIDIA GPU families whether they are running in a desktop, cloud or IoT device. So while developing your application, you do not need to bother about differences between development and deployment platforms.
GPU-accelerated standard BLAS library
GPU-accelerated standard mathematical function library
GPU-accelerated BLAS for sparse matrices
GPU-accelerated random number generation (RNG)
Dense and sparse direct solvers for Computer Vision, CFD, Computational Chemistry, and Linear Optimization applications
GPU accelerated linear solvers for simulations and implicit unstructured methods
GPU-accelerated library for Fast Fourier Transforms
GPU-accelerated library for image and signal processing
High-performance APIs and tools for hardware accelerated video encode and decode
Collective Communications Library for scaling apps across multiple GPUs and nodes
GPU-accelerated library for graph analytics
GPU-accelerated library of parallel algorithms and data structures
GPU-accelerated open-source library for computer vision, image processing and machine learning, now supporting real-time operation
GPU-accelerated open-source Fortran library with functions for math, signal and image processing, statistics, by RogueWave
GPU-accelerated functions for sparse direct solvers, included in SuiteSparse linear algebra package authored by Prof.
GPU-accelerated computational geometry engine for advanced GIS, EDA, computer vision, and motion planning, by Fixstars
Members of the NVIDIA Developer Program get early access to the next CUDA Library release, and access to NVIDIA’s online bug reporting and feature request system.Download Now