You might be in an unfortunate situation then, as the 4060 (Ada), requires a minimum of Cuda 11.8.
Multiple versions of Cuda, side-by-side in the same one environment with Python 3.7 and Ubuntu 22.04
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
CUDA 9.0 ImportError: libcublas.so.8.0 | 17 | 39350 | January 22, 2018 | |
which version of cuda can work with RTX 2080 | 17 | 34213 | May 13, 2021 | |
Cuda Runtime version is insufficent for CUDA runtime version. All of my versions match! | 2 | 734 | October 26, 2018 | |
Installing Cuda in Python virtual environment: environment variables and other | 1 | 1991 | April 18, 2024 | |
Use gpu for tensorflow, crashes | 14 | 6265 | March 7, 2024 | |
[Solved] Tensorflow 1.14 - Cuda 10.0 - GTX 970 - Ubuntu 18.04 | 0 | 2552 | January 27, 2021 | |
[Solved] TensorFlow with GPU in Anaconda env [Ubuntu 16.04 + CUDA 7.5 + cuDNN] | 2 | 44594 | May 24, 2016 | |
ubuntu 18.04.2 CUDA® Toolkit installation use cuda-repo-ubuntu1804-10-1-local-10.1.168-418.67_1.0-1_amd64.deb | 1 | 1567 | August 1, 2019 | |
GTX 1650 No CUDA support | 5 | 8419 | November 10, 2023 | |
Can't get cuda:10.0 docker container to run with tensorflow-gpu | 3 | 1413 | March 4, 2020 |