Instructions to use madebyollin/sdxl-vae-fp16-fix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use madebyollin/sdxl-vae-fp16-fix with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("madebyollin/sdxl-vae-fp16-fix", 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
What is the difference between diffusion_pytorch_model.safetensors and sdxl_vae.safetensors
I went through the readme but did not find an answer.
What is the difference between the file diffusion_pytorch_model.safetensors and the file sdxl_vae.safetensors?
Which one is the VAE fix?
Thanks
They're the same fixed VAE in two different formats (sdxl_vae.safetensors is probably the one you want - the diffusion_pytorch_model is specifically for diffusers iirc).
I tried adding A1111 instructions to the README here https://huggingface.co/madebyollin/sdxl-vae-fp16-fix#automatic1111-usage, though I'm not yet sure if they're correct...
They're the same fixed VAE in two different formats (
sdxl_vae.safetensorsis probably the one you want - thediffusion_pytorch_modelis specifically for diffusers iirc).I tried adding A1111 instructions to the README here https://huggingface.co/madebyollin/sdxl-vae-fp16-fix#automatic1111-usage, though I'm not yet sure if they're correct...
Yet you say to use sdxl.vae.safetensors in the readme
What is the diff between sdxl.vae.safetensors and sdxl_vae.safetensors ??
@NielsGx sdxl.vae.safetensors and sdxl_vae.safetensors are the exact same file (check the hashes), but reportedly automatic1111 preferred the filename sdxl.vae.safetensors