Can RTX 3080 support CUDA 10.1?

Can RTX 3080 support CUDA 10.1 ?
I want to use RTX 3080 for deep learning with RTX 2080 Ti.
Please help me.

I quote this : https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html [CUDA Toolkit v11.1.0][Release Notes]

"Added support for NVIDIA Ampere GPU architecture based GA10x GPUs GPUs (compute capability 8.6), including the GeForce RTX-30 series. "

The answer I believe is no, because support for SM_86 was added in CUDA 11.x. This link might help you understand better : https://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/

If you could point out why you need CUDA 10.1 specifically I might be able to help you.

Thanks for your kindness,jlidiborhen.

I understand that RTX-30 may not support CUDA10 or earlier versions.

The reason why I need CUDA 10.1 is that I’m using RTX-2080 Ti with CUDA 10.1 for a image processing library ,in which different versions of CUDA may not work at the same time.

I’m going to buy new PC.

Thanks.

You can try to compile the library from source with CUDA 11.1, but you should know that TensorRT support for CUDA 11.1 isn’t announced yet (if TRT is one of the dependencies). That way you can use the two GPUs under one version of CUDA (be aware that the motherboard must be SLI Ready).
I could also suggest using docker and dividing the workload, but no docker image supporting RTX 30 is released yet. The only solution I can think of is setting the environment on the host for RTX 3080 and pulling a docker image supporting RTX 2080ti.
Best of luck.