Like I reviewed my YAML and specified a cuda device specifically and I even checked all other things but it won't render off my GPU.
#Low GPU Utilization using Invoke (Windows)
white drift
crisp salmon
You mean in the task Manager is nothing to See at the gpu usage?
Because the graph is not selected that Shows cuda usage. You need to change that to See the usage. (if you mean the task Manager)
white drift
The rendering is taking forever and I'm pretty sure it's not using the card and I can't really validate that its not
I'm just confused as I would figure a card like this wouldn't take tens of minutes to render
crisp salmon
I'm just confused as I would figure a card like this wouldn't take tens of minut...
If you use any flux Model then you should Look at the Low vram Mode https://invoke-ai.github.io/InvokeAI/features/low-vram/ this should speed up Things
white drift
Alright I think that did the trick but now I'm not seeing output in the viewer at all lol and I'm not sure where my output directory is I guess specify that in the yaml file?
No good something is forcibly closing my connection I guess I can't use my GPU to render it seems
crisp salmon
did you maybe check this off?
white drift
I don't know what you are referring to but I just keep getting server errors.... something isn't running right
this doesn't seem self contained or something is preventing it from running right
crisp salmon
did you look at the logs? or post them here so we can find a Solution
white drift
I'm here to ask for help I don't know what the fuck I'm doing
there logs
I don't know what you want
this is just frustratingly stupid
crisp salmon
looks like the CLIPTokenizer cant get be loaded
crisp salmon
?