Instructions to use hakurei/waifu-diffusion with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use hakurei/waifu-diffusion with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("hakurei/waifu-diffusion", 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
- Draw Things
- DiffusionBee
Is the vae included here kl-f8-anime?
#117
by xzuyn - opened
I've been wanting to try the kl-f8-anime vae, but I can only find it in .ckpt and vae.pt format. Somewhere someone listed that it's the waifu diffusion vae, so I'm guessing the safetensors and bin that are included here are kl-f8-anime but named in a more consistent format?
this is same as kl-f8-anime2 (I compared these weights value)
kl-f8-anime2 - 'encoder.conv_in.weight'
tensor([[-0.0163, -0.0705, -0.0105],
[-0.0400, 0.1914, -0.1305],
[-0.0598, 0.1802, 0.0539]], device='cuda:0')
this weight(fp32 bin) - 'encoder.conv_in.weight'
tensor([[-0.0163, -0.0705, -0.0105],
[-0.0400, 0.1914, -0.1305],
[-0.0598, 0.1802, 0.0539]])