Great question — answering as the UphillTech team:
The second error is the useful one: the Docker daemon your client reaches does not have an nvidia runtime registered. Installing the toolkit package alone is not enough, and retrying AlphaFold cannot fix the daemon configuration. Run these from the same shell in which you start AlphaFold: nvidia-smi command -v nvidia-container-runtime docker context show docker info --format '{{json .Runtimes}}' sudo nvidia-ctk runtime configure --runtime=docker sudo systemctl restart docker docker info --format '{{json .Runtimes}}' docker run --rm --gpus all nvidia/cuda:12.4.1-base-ubuntu22.04 nvidia-smi The second docker info must list nvidia . If it does not, you configured a different daemon from the...
Full answer by Андрей: https://superuser.stackexchange.com/a/1940298 (CC BY-SA 4.0).
Need it done for your business? UphillTech.
