Hello,
I have a NVIDIA Quadro RTX 3000 that I want to use for deep learning training.
The website shows this graphics card support CUDA Compute Capability 7.5
Could I use higher cuda version for this?
What is the best torch version should I use?
Thank you
Compute capability is a property of the GPU hardware and immutable for a given GPU. As you already found out, the Quadro RTX 3000 is based on the Turing architecture, with compute capability 7.5 (sm_75).
Support for the Turing architecture was added in CUDA 10.0. The latest CUDA version is 11.2.2. Your GPU will work fine with any CUDA version between those endpoints. However, the various other components in your deep learning software stack may have very specific requirements as to CUDA version, so check relevant information on software requirements for those components.
I do not use PyTorch, but the “Installation” section at the start page for the project (that you linked) seems to indicate that at present, the stable 1.8.1 version of PyTorch requires either CUDA 10.2 or CUDA 11.1. I would not have the faintest clue what the trade-offs are between those two CUDA version with regard to PyTorch. From a general CUDA perspective, choosing the newer version CUDA 11.1 seems appropriate.
Thank you so much njuffa.
Do you mean my graphic card will work fine with CUDA 10.2 and CUDA11.1 although the Nvidia website mentioned RTX3000 is only compatible with 7.5?
This is a misunderstanding. The “7.5” does not refer to a CUDA version. It refers to the compute capability of the GPU. Compute capability indicates the feature set of the GPU architecture. A GPU with compute capability 7.5 requires CUDA version 10 or higher as I mentioned earlier.
It’s been pointed out to me by a diligent reader that the latest CUDA release is now 11.3, which was released on the same day I wrote my previous post, April 15.