greentree's picture
Update README.md
7588682
metadata
license: openrail++
language:
  - en
pipeline_tag: image-to-image
tags:
  - art
  - stable-diffusion-xl
  - onnxruntime-directml
library_name: diffusers

Model Stable Diffusion XL 1.0 Refiner optimized using Microsoft Olive (https://github.com/microsoft/Olive). Provides massively increased generation speed on my AMD RX 7900 XT for images of size 1024*1024. (~10s/it --> ~3.80it/s)

Warning: Requires ONNX Runtime, so this will not work interchangably with any other custom model.

Credit for the original model to StabilityAI: https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0

Other models I've uploaded: https://huggingface.co/greentree/SDXL-olive-optimized, https://huggingface.co/greentree/Dreamshaper-XL-olive-optimized