conda create -n myenv python=3.13
conda activate myenv
pip install 'tensorflow[and-cuda]'
python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"

pip install 'tensorflow[and-cuda]==2.20'
python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"

FIRST THE OUTPUT FOR 2.21
(myenv) [juanma@RScomp ~]$ python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1774864652.884673   70816 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
I0000 00:00:1774864652.906580   70816 cpu_feature_guard.cc:227] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1774864653.421086   70816 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
W0000 00:00:1774864653.546148   70816 gpu_device.cc:2365] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
[]

NOW THE OUTPUT FOR 2.20
(myenv) [juanma@RScomp ~]$ python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
2026-03-30 10:59:31.532594: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-03-30 10:59:31.554544: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-03-30 10:59:32.059641: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]