Chris wrote:
Thanks for the report, and sorry the GPU isn't being picked up. There are a few possible causes, so I'd like to start with five read-only checks. Run them in a Linux terminal (inside WSL if you're on Windows) and paste the results here. 1. Your setup: is this a Linux PC, a Windows PC with WSL, or a NAS? Then run cat /opt/sparkbox/VERSION and send me the number. 2. sudo docker info --format '{{json .Runtimes}}'. Look for "nvidia" in the output. If it's missing, Docker can't pass the card to containers. 3. nvidia-smi. You should see a table with your GPU's name. If you get "command not found" or an error about the driver, the NVIDIA driver isn't ready. 4. sudo sparkbox doctor | grep -i nvidia. This shows which NVIDIA check is failing, if any. 5. grep '^JELLYFINHWACCEL=' /opt/sparkbox/.env. The GPU setting should read JELLYFINHWACCEL=auto. If nothing prints, the setting is missing from this box. Please change nothing yet. Once you send those outputs, I'll tell you the next step.