How to Know CUDA Version
If you are working with CUDA, it is crucial to know the version installed on your system. Here are a few methods to determine the CUDA version.
Method 1: nvcc Compiler
- Open a terminal or command prompt.
- Run the following command:
nvcc --version
.
Example output:
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Sun_Jul_11_22:07:18_PDT_2021
Cuda compilation tools, release 11.4, V11.4.100
Build cuda_11.4.r11.4/compiler.30274561_0
In this example, the CUDA version is 11.4.
Method 2: NVIDIA-SMI Tool
- Open a terminal or command prompt.
- Run the following command:
nvidia-smi
.
Example output (partially shown):
Fri Nov 19 10:21:31 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 465.19.01 Driver Version: 465.19.01 CUDA Version: 11.3 |
|-------------------------------+----------------------+----------------------+
In this example, the CUDA version is 11.3.
Method 3: CUDA Runtime API
- Create a C++ source file, e.g.,
cuda_version.cpp
, using a text editor. - Paste the following code into the file:
“`cpp
#include
#include
int main() {
int cudaVersion;
cudaRuntimeGetVersion(&cudaVersion);
std::cout << “CUDA Version: ” << cudaVersion / 1000 << “.” << cudaVersion % 1000 << std::endl;
return 0;
}
“`
- Save the file and exit the text editor.
- Compile the source file using the CUDA toolkit.
Example command: nvcc cuda_version.cpp -o cuda_version
- Run the compiled program:
./cuda_version
.
Example output:
CUDA Version: 11.4
In this example, the CUDA version is 11.4.
By following these methods, you can easily determine the CUDA version installed on your system.