GPU Utilization
I only have an RTX 2060 Super (8GB).
I am very new to AI models and the like. I have never even used anything else before like SD webui, etc. However, I really liked Animagine-XL-3.0 and tried it on Google Colab again and again! Thank you so much for this awesome work! It was really fun to tinker around with. However, because of the limits on Colab, I opted to also run it locally (despite not knowing anything at all about this).
Fortunately, I was successful in setting things up.
- VS Code
- Downloaded the latest CUDA Toolkit
- PyTorch 2.2.0
I am usually able to generate an image between 4.5 minutes to 6.5 minutes
- Guidance Scale | 7
- Interference Steps | 28
- I have only tried 1024x1024 and 1344x768.
I apologize in advance if this is a rather dumb question but is there any other way to increase utilization of my RTX 2060 Super or is it the best that it can do already? Thank you!
hello, you can try using stable diffusion webui Forge