7 Replies
i keep on getting charged for hours of this not working
From the logs in your screenshots (showing repeated “container is unhealthy: triggered memory limits (OOM)” warnings and 100% memory usage), your pod is crashing because it’s running out of RAM. Even if it’s unresponsive, the container remains active, which is why you’re still being charged. To avoid these issues (and charges), reduce your memory load (e.g., lower batch size or unload models) or upgrade to a plan with more memory… or simply stop the pod if you’re not actively using it.
I’m flagging for our support team to take a closer look.
Looks to be running fine now since you bumped your pod. Was maxing out before.
yeah you can use a higher vram gpu pod
getting this error now for pods that worked fine before using the 4090 on community cloud..is this new?
No, it means the selected specification isn't available
Try other region or specification