User manual and html documentation is available at:

When building the sources yourself you will have to use ILMerge to combine the various DLLs found in the output directory of Cudafy.NET project into one. The new ILMerge GUI (on codeplex) makes this trivial.
Important When opening the Cudafy solution you may get an error from Visual Studio: .nuget\nuget.targets not being found. Right click the solution and click Enable Nuget Package Restore.

Downloading and installing CUDAfy.NET. The installer includes a tool for testing that your system configuration is correct for using CUDAfy.

CUDAfy.NET by Example. A walkthrough of a simple CUDAfy.NET C# program.

Multi-GPUs and Context Switching
Multi-GPUs and context switching is an important topic and will be part of CUDAfy from V1.10 and is as of 20 June 2012 added to SVN. If you are only using one GPU then nothing will change. If you use multiple then this is for you.

Building CUDAfy.NET from sources
A short article explaining how to build CUDAfy.NET from the sources.


Performance Tuning - an excellent series of articles on increasing performance of both CPU and (CUDAfy) GPU code. Importantly it also considers factors such as effort to reward and knowing when to stop! Required reading.

Benchmarking OpenCL and CUDA using CUDAfy on GTX Titan

CUDA Programming Model on AMD GPUs and Intel CPUs

Optimizing Performance of CUDAfy by P. Geerkens

Using Cudafy for GPGPU Programming in .NET

Base64 Encoding on a GPU

High Performance Queries: GPU vs LINQ vs PLINQ

Binomial Option Pricing
This finance sample project is based on the NVIDIA CUDA C sample of the same name. When compared to the CUDA C code it demonstrates just how easy writing CUDAfy applications is. A beta of CUDAfy V1.13 is included. CUDA 5.0 and a GPU with compute 1.3 or greater is required.

Last edited Jan 5 at 10:47 AM by NickKopp, version 21


drewnoakes Jan 15, 2013 at 2:57 PM 
I documented my experience installing the CUDA Toolkit for CUDAfy.NET on Stack Overflow: