Update readme
Browse files
README.md
CHANGED
|
@@ -3,6 +3,7 @@ license: openrail++
|
|
| 3 |
tags:
|
| 4 |
- text-to-image
|
| 5 |
- stable-diffusion
|
|
|
|
| 6 |
---
|
| 7 |
|
| 8 |
# SDXL-Lightning
|
|
@@ -11,7 +12,7 @@ tags:
|
|
| 11 |
|
| 12 |
SDXL-Lightning is a lightning fast text-to-image generative model. It can generate high-quality 1024px images under a few steps. For more information, please refer to our paper: [SDXL-Lightning: Progressive Adversarial Diffusion Distillation](). The models are released for research purposes only.
|
| 13 |
|
| 14 |
-
Our models are distilled from [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0). This repository contains checkpoints for 1-step, 2-step, 4-step, and 8-step distilled models.
|
| 15 |
|
| 16 |
We provide both full UNet and LoRA checkpoints. The full UNet models have the best quality while the LoRA models can be applied to other base models.
|
| 17 |
|
|
@@ -28,7 +29,7 @@ from diffusers import StableDiffusionXLPipeline, EulerDiscreteScheduler
|
|
| 28 |
from huggingface_hub import hf_hub_download
|
| 29 |
|
| 30 |
base = "stabilityai/stable-diffusion-xl-base-1.0"
|
| 31 |
-
repo = "
|
| 32 |
ckpt = "sdxl_lightning_4step_unet.pth" # Use the correct ckpt for your step setting!
|
| 33 |
|
| 34 |
# Load model.
|
|
@@ -50,7 +51,7 @@ from diffusers import StableDiffusionXLPipeline, EulerDiscreteScheduler
|
|
| 50 |
from huggingface_hub import hf_hub_download
|
| 51 |
|
| 52 |
base = "stabilityai/stable-diffusion-xl-base-1.0"
|
| 53 |
-
repo = "
|
| 54 |
ckpt = "sdxl_lightning_4step_lora.pth" # Use the correct ckpt for your step setting!
|
| 55 |
|
| 56 |
# Load model.
|
|
@@ -73,7 +74,7 @@ from diffusers import StableDiffusionXLPipeline, EulerDiscreteScheduler
|
|
| 73 |
from huggingface_hub import hf_hub_download
|
| 74 |
|
| 75 |
base = "stabilityai/stable-diffusion-xl-base-1.0"
|
| 76 |
-
repo = "
|
| 77 |
ckpt = "sdxl_lightning_1step_unet.pth" # Use the correct ckpt for your step setting!
|
| 78 |
|
| 79 |
# Load model.
|
|
|
|
| 3 |
tags:
|
| 4 |
- text-to-image
|
| 5 |
- stable-diffusion
|
| 6 |
+
library_name: diffusers
|
| 7 |
---
|
| 8 |
|
| 9 |
# SDXL-Lightning
|
|
|
|
| 12 |
|
| 13 |
SDXL-Lightning is a lightning fast text-to-image generative model. It can generate high-quality 1024px images under a few steps. For more information, please refer to our paper: [SDXL-Lightning: Progressive Adversarial Diffusion Distillation](). The models are released for research purposes only.
|
| 14 |
|
| 15 |
+
Our models are distilled from [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) (coming soon). This repository contains checkpoints for 1-step, 2-step, 4-step, and 8-step distilled models.
|
| 16 |
|
| 17 |
We provide both full UNet and LoRA checkpoints. The full UNet models have the best quality while the LoRA models can be applied to other base models.
|
| 18 |
|
|
|
|
| 29 |
from huggingface_hub import hf_hub_download
|
| 30 |
|
| 31 |
base = "stabilityai/stable-diffusion-xl-base-1.0"
|
| 32 |
+
repo = "ByteDance/sdxl-lightning"
|
| 33 |
ckpt = "sdxl_lightning_4step_unet.pth" # Use the correct ckpt for your step setting!
|
| 34 |
|
| 35 |
# Load model.
|
|
|
|
| 51 |
from huggingface_hub import hf_hub_download
|
| 52 |
|
| 53 |
base = "stabilityai/stable-diffusion-xl-base-1.0"
|
| 54 |
+
repo = "ByteDance/sdxl-lightning"
|
| 55 |
ckpt = "sdxl_lightning_4step_lora.pth" # Use the correct ckpt for your step setting!
|
| 56 |
|
| 57 |
# Load model.
|
|
|
|
| 74 |
from huggingface_hub import hf_hub_download
|
| 75 |
|
| 76 |
base = "stabilityai/stable-diffusion-xl-base-1.0"
|
| 77 |
+
repo = "ByteDance/sdxl-lightning"
|
| 78 |
ckpt = "sdxl_lightning_1step_unet.pth" # Use the correct ckpt for your step setting!
|
| 79 |
|
| 80 |
# Load model.
|