Nvidia Docker: CUDA driver version is insufficient for CUDA runtime version

Accelerated Computing CUDA CUDA Setup and Installation
1

Running with the following:

My Dockerfile:

FROM nvidia/cuda:10.1-devel-ubuntu18.04

ENV DEBIAN_FRONTEND=noninteractive
WORKDIR /opt/memtest
RUN apt-get update \
 && apt-get install -y --no-install-recommends \
        curl \
 && rm -rf /var/lib/apt/lists/* \
 && curl -s -S -L https://github.com/ForkLab/cuda_memtest/archive/refs/heads/dev.tar.gz \
 		| tar -xzf - --strip-components=1
RUN make CFLAGS="-arch compute_50 -DSM_50 -O3"

Built with sudo docker build -t memtest .

Output of sudo docker run --gpus all --rm -it memtest nvidia-smi:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.211.00   Driver Version: 418.211.00   CUDA Version: N/A      |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Quadro K620         On   | 00000000:01:00.0 Off |                  N/A |
| 34%   39C    P8     1W /  30W |      1MiB /  2002MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

Output of sudo docker run --gpus all --rm -it memtest ./cuda_memtest:

[10/16/2021 16:44:24][411cc62bdc89][0]:Running cuda memtest, version 1.2.3
[10/16/2021 16:44:24][411cc62bdc89][0]:ERROR: CUDA error: CUDA driver version is insufficient for CUDA runtime version, line 283, file cuda_memtest.cu
[10/16/2021 16:44:24][411cc62bdc89][0]:ERROR: CUDA error: CUDA driver version is insufficient for CUDA runtime version, line 283, file cuda_memtest.cu

I don’t understand why this error is happening? I believe the K620 has CUDA compute capability 5.0 which should be sufficient, and my driver 418.211 should be recent enough for the cuda:10.1-devel-ubuntu18.04 image.