CUDA Toolkit 4.2
| CUDA FOR WINDOWS | ||||||||||||||||||||
|
|
|
|
||||||||||||||||||
| CUDA FOR LINUX | ||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||
![]() |
Download Drivers(ver 295.41)![]() ![]() |
![]() |
Download SDK![]() |
|||||||||||||||||||||||||||||||
| CUDA FOR MAC | ||
|
|
|
|
This release of the CUDA Toolkit enables development using GPUs using the Kepler architecture, such as the GeForce GTX680. Feature and functionality builds on the foundation of the CUDA 4.1 release which introduced:
- A new LLVM-based CUDA compiler
- 1000+ new image processing functions
- Redesigned Visual Profiler with automated performance analysis and integrated expert guidance
A new version of Nsight Visual Studio Edition 2.2 has also been released which includes support for CUDA 4.2 and Kepler, find out more.
To report issues and bugs - please sign up as a CUDA Registered Developer. To get a quick overview of features and performance -download the CUDA 4.x feature overview presentation
| Useful Links |
Getting Started Guide Documentation |
CUDA 4.x Overview Webinar Tools and Libraries Ecosystem |
Become a CUDA Registered Developer CUDA Webinars |
Learn more about whats included in the CUDA Toolkit and GPU Computing SDK .
Get the latest developer CUDA insights by attending CUDA Training Webinars
Find out about new technologies such as GPUDirect, which are eliminating bottlenecks and making parallel computing easier than ever before.
If you need a legacy version of the CUDA Toolkit please visit our CUDA Toolkit Archive
Download

Download
Download



Registered Developers Website
NVDeveloper (old site)