File size: 743 Bytes
ef0b7c9
 
e3e718f
 
4c34c58
4604249
 
 
7588682
7c2e593
7212725
f33a943
7212725
83c4932
a474118
 
3269b88
0230499
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
---
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