#why no matter stop/start or restart directly. the GPU VRAM take almost 85%?
1 messages · Page 1 of 1 (latest)
this is the startup status
and I migrate to another new pod, the problem is still. how to fix this.
well, it kind of make sense that it is the same after migration. There might be something scheduled to run after startup or several other problems
try to run these:
which nvidia-smi nvidia-smi
also look into logs
and tell me which template are you using
Ok I will do this migration again. It may need some time
sure, let me know once it finishes