Instructions to use Vaelico/Wulver with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Vaelico/Wulver with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Vaelico/Wulver", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
stretched-out-glitch on some images
On some images, there is a glitch on the upper or lower edge (see attached image crop) that appears in the Q8 and BF16 variants of Wulver 0.5.
(image from: https://civitai.com/images/143972444)
If this was already a known issue: sorry, I didn't find this!
I hope this is not too late and that this issue can be easily fixed as training continues, so v1.0 will be perfect (Keep up the good work!).
PS: This glitch also appears when generating images with Vaelico (https://app.vaelico.ai/).
Yes, traing issue, resolved, it was a dumb little thign that contaminated so much, is simply random and can't be helped. I will update it here too the issue. thanks.
Oh, and in the web i simply fixed by croping, is slightly more expensive. like 6%.
Fingers crossed that further training will completely solve/fix this; very excited for v1.0!