RTX 3070, Cuda and Tensorflow version

Hi,
I have cuda 10.1 version installed (and Nvidia RTX 3070 in an egpu) and following all the step in https://developer.nvidia.com/blog/accelerating-machine-learning-on-a-linux-laptop-with-an-external-gpu/

I used tensorflow as:

import tensorflow as tf
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices(‘GPU’)))

Num GPUs Available: 1

But when I put:

tf.test.is_gpu_available()

RuntimeError: CUDA runtime implicit initialization on GPU:0 failed. Status: device kernel image is invalid

The problem is de cuda version??
the card work with a low version of cuda (10.1)??

Thanks for all your help!