How to check cuda version
How to check cuda version
How to check which CUDA version is installed on Linux
There are several ways and steps you could check which CUDA version is installed on your Linux box.
Check if CUDA is installed and it’s location with NVCC
Run which nvcc to find if nvcc is installed properly.
You should see something like /usr/bin/nvcc. If that appears, your NVCC is installed in the standard directory.
If you have installed the CUDA toolkit but which nvcc returns no results, you might need to add the directory to your path.
This means that we have CUDA version 8.0.61 installed.
Get CUDA version from CUDA code
When you’re writing your own code, figuring out how to check the CUDA version, including capabilities is often accomplished with the cudaDriverGetVersion () API call.
The API call gets the CUDA version from the active driver, currently loaded in Linux or Windows.
Identifying which CUDA driver version is installed and active in the kernel
You can also use the kernel to run a CUDA version check:
Identifying which GPU card is installed and what version
In many cases, I just use nvidia-smi to check the CUDA version on CentOS and Ubuntu.
For me, nvidia-smi is the most straight-forward and simplest way to get a holistic view of everything – both GPU card model and driver version, as well as some additional information like the topology of the cards on the PCIe bus, temperatures, memory utilization, and more.
The driver version is 367.48 as seen below, and the cards are two Tesla K40m.
Troubleshooting
After installing a new version of CUDA, there are some situations that require rebooting the machine to have the driver versions load properly. It is my recommendation to reboot after performing the kernel-headers upgrade/install process, and after installing CUDA – to verify that everything is loaded correctly.
How to Check CUDA Version Easily
Here you will learn how to check NVIDIA CUDA version in 3 ways: nvcc from CUDA toolkit, nvidia-smi from NVIDIA driver, and simply checking a file. Using one of these methods, you will be able to see the CUDA version regardless the software you are using, such as PyTorch, TensorFlow, conda (Miniconda/Anaconda) or inside docker.
» class=»wp_ulike_btn wp_ulike_put_image wp_post_btn_2895″>
Prerequisite
What is CUDA?
CUDA is a general parallel computing architecture and programming model developed by NVIDIA for its graphics cards (GPUs). Using CUDA, PyTorch or TensorFlow developers will dramatically increase the performance of PyTorch or TensorFlow training models, utilizing GPU resources effectively.
Method 1 — Use nvcc to check CUDA version
To check CUDA version with nvcc, run
You can see similar output in the screenshot below. The last line shows you version of CUDA. The version here is 10.1. Yours may vary, and can be either 10.0, 10.1, 10.2 or even older versions such as 9.0, 9.1 and 9.2. After the screenshot you will find the full text output too.
What is nvcc?
Check out nvcc ‘s manpage for more information.
Method 2 — Check CUDA version by nvidia-smi from NVIDIA Linux driver
The second way to check CUDA version is to run nvidia-smi, which comes from downloading the NVIDIA driver, specifically the NVIDIA-utils package. You can install either Nvidia driver from the official repositories of Ubuntu, or from the NVIDIA website.
You can see similar output in the screenshot below. The version is at the top right of the output. Here’s my version is CUDA 10.2. You may have 10.0, 10.1 or even the older version 9.0 or 9.1 or 9.2 installed.
Importantly, except for CUDA version. There are more details in the nvidia-smi output, driver version (440.100), GPU name, GPU fan percentage, power consumption/capability, memory usage, can also be found here. You can also find the processes which use the GPU at the moment. This is helpful if you want to see if your model or system is using GPU such as PyTorch or TensorFlow.
Here is the full text output:
What is nvidia-smi?
nvidia-smi (NVSMI) is NVIDIA System Management Interface program. It is also known as NVSMI. nvidia-smi provides monitoring and maintenance capabilities for all of tje Fermi’s Tesla, Quadro, GRID and GeForce NVIDIA GPUs and higher architecture families. For most functions, GeForce Titan Series products are supported with only little detail given for the rest of the Geforce range.
NVSMI is also a cross-platform application that supports both common NVIDIA driver-supported Linux distros and 64-bit versions of Windows starting with Windows Server 2008 R2. Metrics may be used directly by users via stdout, or stored via CSV and XML formats for scripting purposes.
Method 3 — cat /usr/local/cuda/version.txt
Note that if you install Nvidia driver and CUDA from Ubuntu 20.04’s own official repository this approach may not work.
3 ways to check CUDA version
Time Needed : 5 minutes
There are basically three ways to check CUDA version. One must work if not the other.
Run cat /usr/local/cuda/version.txt
Note: this may not work on Ubuntu 20.04
How to Check CUDA Version on Ubuntu 18.04
» class=»wp_ulike_btn wp_ulike_put_image wp_post_btn_3132″>
Prerequisite
Before we start, you should have installed NVIDIA driver on your system as well as Nvidia CUDA toolkit.
Method 1 — Use nvidia-smi from Nvidia Linux driver
The first way to check CUDA version is to run nvidia-smi that comes from your Ubuntu 18.04’s NVIDIA driver, specifically the NVIDIA-utils package. You can install either Nvidia driver from the official repository of Ubuntu, or from the NVIDIA website.
To use nvidia-smi to check your CUDA version on Ubuntu 18.04, directly run from command line
You will see similar output to the screenshot below. The details about the CUDA version is to the top right of the output. My version is 10.2 here. Whether you have 10.0, 10.1 or even the older 9.0 installed, it will differ.
Surprisingly, except for the CUDA version, you can also find more detail from nvidia-smi, such as driver version (440.64), GPU name, GPU fan ratio, power consumption / capacity, memory usage. Also you can find the processes that actually use the GPU.
Here is the full text output:
What is nvidia-smi?
nvidia-smi (NVSMI) is NVIDIA System Management Interface program. It is also known as NVSMI. nvidia-smi provides tracking and maintenance features for all of the Tesla, Quadro, GRID and GeForce NVIDIA GPUs and higher architectural families in Fermi. For most functions, GeForce Titan Series products are supported with only a limited amount of detail provided for the rest of the Geforce range.
NVSMI is also a cross-platform program which supports all common NVIDIA driver-supported Linux distros and 64-bit versions of Windows starting with Windows Server 2008 R2. Metrics can be used by users directly via stdout, or saved in CSV and XML formats for scripting purposes.
For more information, check out nvidia-smi ‘s manpage.
Method 2 — Use nvcc to check CUDA version on Ubuntu 18.04
To check the CUDA version with nvcc on Ubuntu 18.04, execute
Different output can be seen in the screenshot below. The last line reveals a version of your CUDA version. This version here is 10.1. Yours may vary, and may be 10.0 or 10.2. You will see the full text output after the screenshot too.
What is nvcc?
Check out the manpage of nvcc for more information.
Method 3 — cat /usr/local/cuda/version.txt
Note that this method might not work on Ubuntu 18.04 if you install Nvidia driver and CUDA from Ubuntu 18.04’s own official repository.
3 ways to check CUDA version on Ubuntu 18.04
Time Needed : 5 minutes
There are three ways to identify the CUDA version on Ubuntu 18.04.
Simply run nvidia-smi
Run cat /usr/local/cuda/version.txt 
nvcc –version
bycloudai/SwapCudaVersionWindows
Use Git or checkout with SVN using the web URL.
Work fast with our official CLI. Learn more.
Launching GitHub Desktop
If nothing happens, download GitHub Desktop and try again.
Launching GitHub Desktop
If nothing happens, download GitHub Desktop and try again.
Launching Xcode
If nothing happens, download Xcode and try again.
Launching Visual Studio Code
Your codespace will open once ready.
There was a problem preparing your codespace, please try again.
Latest commit
Git stats
Files
Failed to load latest commit information.
README.md
Swap CUDA Toolkit Versions on Windows
Here I will do a quick run down on how to swap CUDA versions.
For ease, I will be demonstrating switching from CUDA 11.6 to CUDA 11.3, the same methods apply to other versions.
Step 0: Check CUDA Version
Check what version of CUDA you have. You can enter this to any command prompt (cmd, anaconda, etc)
if you get something like this:
This means you don’t have any CUDA installed. You can download your desired CUDA Toolkit version here (everything default would be fine)
A quick rule of thumb:
You can also check your GPU compatibility here for NVIDIA GPU = 3.7, then it supports PyTorch.
if you get something like this:
Then it means you have CUDA installed. And in my case, it’s CUDA 11.6. I will be swapping to CUDA 11.3 in the following steps.
Step 1: Locate System Environment Variables
Open up your environment variables. You can search «env» in the search tab, it should look something like this.
Then open it. Then click «Environment Variables»
Then it should open up a winodw like this
Step 2: Change System Variables
Double check on CUDA_PATH and this window should pop up
Then enter the target version of your CUDA there. In my case it’s changing 11.6 to 11.3
Press ok and proceed next step.
Step 3: Change System Paths
You should see your current version on the very top. You going have to move your desired version to the very top
So it should look like this after moving
Press ok and you may now close all the windows for environment variables & system properties.
Step 4: Check if succeed
Close the last command prompt, and open a new one. Enter the following command:
If it outputs your desired version, then you have succeed in swapping CUDA version.
About
How to swap/switch CUDA versions on Windows
How to check cuda version
The installation instructions for the CUDA Toolkit on MS-Windows systems.
1. Introduction
CUDA В® is a parallel computing platform and programming model invented by NVIDIA. It enables dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU).
This guide will show you how to install and check the correct operation of the CUDA development tools.
1.1. System Requirements
The next two tables list the currently supported Windows operating systems and compilers.
| Operating System | Native x86_64 | Cross (x86_32 on x86_64) |
|---|---|---|
| Windows 11 | YES | NO |
| Windows 10 | YES | NO |
| Windows Server 2022 | YES | NO |
| Windows Server 2019 | YES | NO |
| Windows Server 2016 | YES | NO |
| Compiler* | IDE | Native x86_64 | Cross (x86_32 on x86_64) |
|---|---|---|---|
| MSVC Version 193x | Visual Studio 2022 17.0 | YES | YES |
| MSVC Version 192x | Visual Studio 2019 16.x | YES | YES |
| MSVC Version 191x | Visual Studio 2017 15.x (RTW and all updates) | YES | YES |
* Support for Visual Studio 2015 is deprecated in release 11.1.
x86_32 support is limited. See the x86 32-bit Support section for details.
For more information on MSVC versions, Visual Studio product versions, visit https://dev.to/yumetodo/list-of-mscver-and-mscfullver-8nd.
1.2. x86 32-bit Support
Native development using the CUDA Toolkit on x86_32 is unsupported. Deployment and execution of CUDA applications on x86_32 is still supported, but is limited to use with GeForce GPUs. To create 32-bit CUDA applications, use the cross-development capabilities of the CUDA Toolkit on x86_64.
1.3. About This Document
This document is intended for readers familiar with Microsoft Windows operating systems and the Microsoft Visual Studio environment. You do not need previous experience with CUDA or experience with parallel computation.
2. Installing CUDA Development Tools
Basic instructions can be found in the Quick Start Guide. Read on for more detailed instructions.
2.1. Verify You Have a CUDA-Capable GPU
You can verify that you have a CUDA-capable GPU through the Display Adapters section in the Windows Device Manager. Here you will find the vendor name and model of your graphics card(s). If you have an NVIDIA card that is listed in http://developer.nvidia.com/cuda-gpus, that GPU is CUDA-capable. The Release Notes for the CUDA Toolkit also contain a list of supported products.
2.2. Download the NVIDIA CUDA Toolkit
The CUDA Toolkit installs the CUDA driver and tools needed to create, build and run a CUDA application as well as libraries, header files, and other resources.
Download Verification
The download can be verified by comparing the MD5 checksum posted at https://developer.download.nvidia.com/compute/cuda/11.6.2/docs/sidebar/md5sum.txt with that of the downloaded file. If either of the checksums differ, the downloaded file is corrupt and needs to be downloaded again.
To calculate the MD5 checksum of the downloaded file, follow the instructions at https://support.microsoft.com/kb/889768.
2.3. Install the CUDA Software
Graphical Installation
Install the CUDA Software by executing the CUDA installer and following the on-screen prompts.
Silent Installation
| Subpackage Name | Subpackage Description |
|---|---|
| Toolkit Subpackages (defaults to C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v 11.7 ) | |
| cudart_ 11.7 | CUDA Runtime libraries. |
| cuobjdump_ 11.7 | Extracts information from cubin files. |
| cupti_ 11.7 | The CUDA Profiling Tools Interface for creating profiling and tracing tools that target CUDA applications. |
| cuxxfilt_ 11.7 | The CUDA cu++ filt demangler tool. |
| demo_suite_ 11.7 | Prebuilt demo applications using CUDA. |
| documentation_ 11.7 | CUDA HTML and PDF documentation files including the CUDA C++ Programming Guide, CUDA C++ Best Practices Guide, CUDA library documentation, etc. |
| memcheck_ 11.7 | Functional correctness checking suite. |
| nvcc_ 11.7 | CUDA compiler. |
| nvdisasm_ 11.7 | Extracts information from standalone cubin files. |
| nvml_dev_ 11.7 | NVML development libraries and headers. |
| nvprof_ 11.7 | Tool for collecting and viewing CUDA application profiling data from the command-line. |
| nvprune_ 11.7 | Prunes host object files and libraries to only contain device code for the specified targets. |
| nvrtc_ 11.7 | |
Extracting and Inspecting the Files Manually
Sometimes it may be desirable to extract or inspect the installable files directly, such as in enterprise deployment, or to browse the files before installation. The full installation package can be extracted using a decompression tool which supports the LZMA compression method, such as 7-zip or WinZip.
2.3.1. Uninstalling the CUDA Software
All subpackages can be uninstalled through the Windows Control Panel by using the Programs and Features widget.
2.4. Using Conda to Install the CUDA Software
This section describes the installation and configuration of CUDA when using the Conda installer. The Conda packages are available at https://anaconda.org/nvidia.
2.4.1. Conda Overview
2.4.2. Installation
To perform a basic install of all CUDA Toolkit components using Conda, run the following command:
2.4.3. Uninstallation
To uninstall the CUDA Toolkit using Conda, run the following command:
2.4.4. Installing Previous CUDA Releases
All Conda packages released under a specific CUDA version are labeled with that release version. To install a previous version, include that label in the install command such as:
Some CUDA releases do not move to new versions of all installable components. When this is the case these components will be moved to the new label, and you may need to modify the install command to include both labels such as:
This example will install all packages released as part of CUDA 11.3.0.
2.5. Use a Suitable Driver Model
On Windows 10 and later, the operating system provides two under which the NVIDIA Driver may operate:
2.6. Verify the Installation
Before continuing, it is important to verify that the CUDA toolkit can find and communicate correctly with the CUDA-capable hardware. To do this, you need to compile and run some of the included sample programs.
2.6.1. Running the Compiled Examples
Start > All Programs > Accessories > Command Prompt
CUDA Samples are located in https://github.com/nvidia/cuda-samples. To use the samples, clone the project, build the samples, and run them using the instructions on the Github page.
To verify a correct configuration of the hardware and software, it is highly recommended that you build and run the deviceQuery sample program. The sample can be built using the provided VS solution files in the deviceQuery folder.
This assumes that you used the default installation directory structure. If CUDA is installed and configured correctly, the output should look similar to Figure 1.
The exact appearance and the output lines might be different on your system. The important outcomes are that a device was found, that the device(s) match what is installed in your system, and that the test passed.
If a CUDA-capable device and the CUDA Driver are installed but deviceQuery reports that no CUDA-capable devices are present, ensure the deivce and driver are properly installed.
Running the bandwidthTest program, located in the same directory as deviceQuery above, ensures that the system and the CUDA-capable device are able to communicate correctly. The output should resemble Figure 2.
The device name (second line) and the bandwidth numbers vary from system to system. The important items are the second line, which confirms a CUDA device was found, and the second-to-last line, which confirms that all necessary tests passed.
If the tests do not pass, make sure you do have a CUDA-capable NVIDIA GPU on your system and make sure it is properly installed.
3. Pip Wheels
NVIDIA provides Python Wheels for installing CUDA through pip, primarily for using CUDA with Python. These packages are intended for runtime use and do not currently include developer tools (these can be installed separately).
Please note that with this installation method, CUDA installation environment is managed via pip and additional care must be taken to set up your host environment to use CUDA outside the pip environment.
4. Compiling CUDA Programs
4.1. Compiling Sample Projects
The bandwidthTest project is a good sample project to build and run. It is located in https://github.com/NVIDIA/cuda-samples/tree/master/Samples/bandwidthTest.
4.2. Sample Projects
The sample projects come in two configurations: debug and release (where release contains no debugging information) and different Visual Studio projects.
A few of the example projects require some additional setup.
4.3. Build Customizations for New Projects
When creating a new CUDA application, the Visual Studio project file must be configured to include CUDA build customizations. To accomplish this, click File-> New | Project. NVIDIA-> CUDA->, then select a template for your CUDA Toolkit version. For example, selecting the «CUDA 11.7 Runtime» template will configure your project for use with the CUDA 11.7 Toolkit. The new project is technically a C++ project (.vcxproj) that is preconfigured to use NVIDIA’s Build Customizations. All standard capabilities of Visual Studio C++ projects will be available.
4.4. Build Customizations for Existing Projects
While Option 2 will allow your project to automatically use any new CUDA Toolkit version you may install in the future, selecting the toolkit version explicitly as in Option 1 is often better in practice, because if there are new CUDA configuration options added to the build customization rules accompanying the newer toolkit, you would not see those new options using Option 2.
5. Additional Considerations
A number of helpful development tools are included in the CUDA Toolkit or are available for download from the NVIDIA Developer Zone to assist you as you develop your CUDA programs, such as NVIDIA В® Nsightв„ў Visual Studio Edition, NVIDIA Visual Profiler, and cuda-memcheck.
For technical support on programming questions, consult and participate in the developer forums at http://developer.nvidia.com/cuda/.
Notices
Notice
This document is provided for information purposes only and shall not be regarded as a warranty of a certain functionality, condition, or quality of a product. NVIDIA Corporation (“NVIDIA”) makes no representations or warranties, expressed or implied, as to the accuracy or completeness of the information contained in this document and assumes no responsibility for any errors contained herein. NVIDIA shall have no liability for the consequences or use of such information or for any infringement of patents or other rights of third parties that may result from its use. This document is not a commitment to develop, release, or deliver any Material (defined below), code, or functionality.
NVIDIA reserves the right to make corrections, modifications, enhancements, improvements, and any other changes to this document, at any time without notice.
Customer should obtain the latest relevant information before placing orders and should verify that such information is current and complete.
NVIDIA products are sold subject to the NVIDIA standard terms and conditions of sale supplied at the time of order acknowledgement, unless otherwise agreed in an individual sales agreement signed by authorized representatives of NVIDIA and customer (“Terms of Sale”). NVIDIA hereby expressly objects to applying any customer general terms and conditions with regards to the purchase of the NVIDIA product referenced in this document. No contractual obligations are formed either directly or indirectly by this document.
NVIDIA products are not designed, authorized, or warranted to be suitable for use in medical, military, aircraft, space, or life support equipment, nor in applications where failure or malfunction of the NVIDIA product can reasonably be expected to result in personal injury, death, or property or environmental damage. NVIDIA accepts no liability for inclusion and/or use of NVIDIA products in such equipment or applications and therefore such inclusion and/or use is at customer’s own risk.
NVIDIA makes no representation or warranty that products based on this document will be suitable for any specified use. Testing of all parameters of each product is not necessarily performed by NVIDIA. It is customer’s sole responsibility to evaluate and determine the applicability of any information contained in this document, ensure the product is suitable and fit for the application planned by customer, and perform the necessary testing for the application in order to avoid a default of the application or the product. Weaknesses in customer’s product designs may affect the quality and reliability of the NVIDIA product and may result in additional or different conditions and/or requirements beyond those contained in this document. NVIDIA accepts no liability related to any default, damage, costs, or problem which may be based on or attributable to: (i) the use of the NVIDIA product in any manner that is contrary to this document or (ii) customer product designs.
No license, either expressed or implied, is granted under any NVIDIA patent right, copyright, or other NVIDIA intellectual property right under this document. Information published by NVIDIA regarding third-party products or services does not constitute a license from NVIDIA to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property rights of the third party, or a license from NVIDIA under the patents or other intellectual property rights of NVIDIA.
Reproduction of information in this document is permissible only if approved in advance by NVIDIA in writing, reproduced without alteration and in full compliance with all applicable export laws and regulations, and accompanied by all associated conditions, limitations, and notices.
THIS DOCUMENT AND ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE. TO THE EXTENT NOT PROHIBITED BY LAW, IN NO EVENT WILL NVIDIA BE LIABLE FOR ANY DAMAGES, INCLUDING WITHOUT LIMITATION ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, HOWEVER CAUSED AND REGARDLESS OF THE THEORY OF LIABILITY, ARISING OUT OF ANY USE OF THIS DOCUMENT, EVEN IF NVIDIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Notwithstanding any damages that customer might incur for any reason whatsoever, NVIDIA’s aggregate and cumulative liability towards customer for the products described herein shall be limited in accordance with the Terms of Sale for the product.