Develop cross-platform NVIDIA® CUDA® solutions with CUDA.NET.
CUDA.NET is an effort to provide access to CUDA® functionality for .NET applications.

Using GPU acceleration and CUDA.NET, it is possible to achieve great performance in .NET based applications, offloading CPU computations to the GPU – a dedicated and standardized hardware.
Real-time performance has never been so easy, even in .NET applications, as the GPU can process massive amounts of data at a fraction of a second.

Enjoy the advantage of CUDA.NET for cross-platform solution – once the application is ready, it can be used under Windows, Linux, MacOSX in 32 or 64 bit editions, supporting further CUDA® compliant environments, transparently.


CUDA.NET is provided free of charge, for private, academic, research or commercial use.
The library is provided as is, without warranty or support of any kind.
For support or development services:


Checkout CUDA.NET archive page.


  • Double precision
  • Object oriented model for CUDA programming
  • NVIDIA® CUDA® 3.0 Driver API
  • NVIDIA® CUDA® 3.0 Runtime API
  • CUFFT routines
  • CUBLAS routines
  • Direct3D interoperability
  • OpenGL interoperability
  • Synchronization capabilities for multi-device/multi-threaded applications

Operating System Support

  • Microsoft® Windows®
  • Linux 32/64 bit (using Mono)
  • MacOSX (using Mono)

Requirements: NVIDIA® CUDA® version 1.1 and up, .NET Framework 2.0 or newer.