Release Highlights
The Nvidia CUDA toolkit is an extension of the GPU parallel computing platform and programming model. The Nvidia CUDA installation consists of inclusion of the official Nvidia CUDA repository followed by the installation of relevant meta package and configuring path the the executable CUDA binaries. GPUDirect RDMA is a technology introduced in Kepler-class GPUs and CUDA 5.0 that enables a direct path for data exchange between the GPU and a third-party peer device using standard features of PCI Express.
Easier Application Porting
- Share GPUs across multiple threads
- Use all GPUs in the system concurrently from a single host thread
- No-copy pinning of system memory, a faster alternative to cudaMallocHost()
- C++ new/delete and support for virtual functions
- Support for inline PTX assembly
- Thrust library of templated performance primitives such as sort, reduce, etc.
- NVIDIA Performance Primitives (NPP) library for image/video processing
- Layered Textures for working with same size/format textures at larger sizes and higher performance
Faster Multi-GPU Programming
- Unified Virtual Addressing
- GPUDirect v2.0 support for Peer-to-Peer Communication
New & Improved Developer Tools
- Automated Performance Analysis in Visual Profiler
- C++ debugging in CUDA-GDB for Linux and MacOS
- GPU binary disassembler for Fermi architecture (cuobjdump)
- Parallel Nsight 2.0 now available for Windows developers with new debugging and profiling features.
Watch the CUDA Toolkit 4.0 Feature and Overview Webinar (or just the slides) for an overview of some of the exciting new features of this release.
Check out the NEW CUDA 4.0 Math Library Performance Review
Nvidia Cuda Toolkit 9
Find all the latest versions of other Libraries and Tools on our Tools & EcoSystem Page
Please download the lastest CUDA Toolkit 4.0 Errata Update.
The latest released NVIDIA Drivers are always available at www.nvidia.com/drivers
For previous releases, see the CUDA Toolkit Release Archive
Get yourself fully trained- check out the latest CUDA Webinars
Become a CUDA Registered Developer, report bugs, engage with NVIDIA engineering
Jump to: [Windows][ Linux ] [ MacOS ]
| Windows 7, VISTA, Windows XP | Downloads |
|---|---|
| Developer Drivers for WinXP (270.81) Support for XP on notebooks is being phased out and is not available for this release. See Release Notes and Getting Started Guides for more information. | |
| Developer Drivers for WinVista and Win7 (270.81) | |
| Notebook Developer Drivers for WinVista and Win7 (275.33) | |
CUDA Toolkit
| |
| *NEW* CUDA Toolkit 4.0 Build Customization BUG FIX Update Fixes error message '$(CUDABuildTasksPath) property is not valid' | download |
| GPU Computing SDK - complete package including all code samples | 32-bit64-bit browse online |
| Parallel Nsight 2.0 | download |
| Learn about additional tools, libraries, and more… | CUDA Ecosystem |
| CUDA Tools SDK (APIs for 3rd party performance analysis tools and cluster management solutions) |
| Linux | Downloads |
|---|---|
| Developer Drivers for Linux (270.41.19) | |
CUDA Toolkit
| |
| CUDA Toolkit for Fedora 13 | 32-bit, (Visual Profiler_Patch) 64-bit, (Visual Profiler Patch) |
| CUDA Toolkit for RedHat Enterprise Linux 6.0 | 64-bit, (Visual Profiler Patch) |
| CUDA Toolkit for RedHat Enterprise Linux 5.5 | 32-bit, (Visual Profiler Patch) 64-bit, (Visual Profiler Patch) |
| CUDA Toolkit for RedHat Enterprise Linux 4.8 | |
| CUDA Toolkit for Ubuntu Linux 10.10 | 32-bit, (Visual Profiler Patch) 64-bit, (Visual Profiler Patch) |
| CUDA Toolkit for OpenSUSE 11.2 | 32-bit, (Visual Profiler Patch) 64-bit, (VP Patch -coming soon) |
| CUDA Toolkit for SUSE Linux Enterprise Server 11 SP1 | 32-bit, (Visual Profiler Patch) 64-bit, (Visual Profiler Patch) |
| GPU Computing SDK - complete package including all code samples | download browse online |
| Learn about additional tools, libraries, and more… | CUDA Ecosystem |
| CUDA Tools SDK (APIs for 3rd party debuggers, performance analysis tools and cluster management solutions) |
Nvidia Cuda Toolkit 11.0
| Mac OS X | Downloads |
|---|---|
| Developer Drivers (4.0.50) for MacOS (requires OS ver. 10.6.8 or higher) | download |
CUDA Toolkit (requires OS version 10.6.7 or higher)
| |
| GPU Computing SDK - complete package including all code samples | download Browse Online |
| Learn about additional tools, libraries, and more… | CUDA Ecosystem |
| CUDA Tools SDK (APIs for 3rd party debuggers and performance analysis tools) | download |