How can I run tensorflow on a desktop RTX 3050 graphics card?

Can the RTX 3050 desktop graphics card be used with “its GPU computing power” with tensorflow? since it does not appear in this link CUDA GPUs - Compute Capability | NVIDIA Developer, can I use the drivers of the portable version of the same graphics card to be able to work with it? How can I do it so that it can work with this limitation?

First off, make sure your desktop has the latest NVIDIA drivers installed. Then, grab the TensorFlow GPU version either through pip or Anaconda. Lastly, don’t forget to set up CUDA and cuDNN libraries to tap into that sweet GPU power for faster machine learning magic!

Regard: Sunder Sen
“Modified by moderator”

Is there a tutorial to configure the CUDA and cuDNN libraries, how to use it with anaconda?

Hi @Stefano_Tabarroni, As RTX 3050 has compute capability 8.6 it can support CUDA 12.x version. Miniconda is the recommended approach for installing TensorFlow with GPU support and also the easiest way to install the required software especially for the GPU setup. Once you have installed miniconda you create a new virtual environment using the cond create command. For example,

conda create --name tf python=3.9

Then you can activate the created environment using

conda activate tf

Then you can install Tensorflow and CUDA libraries using

python3 -m pip install tensorflow[and-cuda]==2.15.0

Thank You.