For one of my project, I wish to query if a NVIDIA gpu is compute capable without linking to CUDA apis. Is there any other extensions that would help me achieve this? I am not ok to use nvml as this is also part of the CUDA toolkit.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| pyNVML and nvmlDeviceGetCudaComputeCapability() | 1 | 1947 | March 23, 2019 | |
| Retrieving support Compute Capabilities of the installed nvcc compiler from CMake | 0 | 565 | November 15, 2018 | |
| What about compute capabilities of devices ehich are not listed in Appendix A of the Programming Gui | 2 | 4158 | August 12, 2009 | |
| Why NVML API can get the GPU but the CUDA API can't | 2 | 911 | October 12, 2021 | |
| Why 1050 Notebook doesn't support CUDA? | 1 | 667 | January 13, 2017 | |
| Determining supported compute capabilities | 3 | 1212 | January 25, 2024 | |
| Can the program compiled with NVCC run a machine without the GPU card? | 15 | 10959 | June 15, 2011 | |
| No Cuda Capable devices are detected in Nvidia Titan Xp | 3 | 1757 | October 14, 2021 | |
| Compute processes: not supported | 4 | 24491 | March 3, 2014 | |
| detecting if a gpu is driving a display | 6 | 2476 | February 24, 2009 |