CUDA Toolkit

Provides a comprehensive environment for C/C++ developers building GPU-accelerated applications.


Directives for parallel computing, is a new open parallel programming standard designed to enable all scientific and technical programmers.

PGI Accelerator Fortran and C Compilers

Accelerate applications on GPU platforms by adding compiler directives to existing code.

The PGI CUDA C/C++ compiler for x86

Compile and optimize their CUDA applications to run on x86-based workstations, servers and clusters.


Enjoy GPU acceleration directly from your Fortran program using CUDA Fortran from The Portland Group.

Anaconda Accelerate

Enables acceleration on your GPU or multi-core processor using Python.


Gives you access to CUDA fuctionality from your Python code.

Altimesh Hybridizer

An advanced productivity tool that generates vectorized C++ (AVX) and CUDA C code from .NET assemblies (MSIL) or Java archives (bytecode)


OpenCL is a low-level API for GPU computing that can run on CUDA-powered GPUs.

Alea GPU

This is a novel approach to develop GPU applications on .NET, combining the CUDA with Microsoft’s F#.