Hugging Face's logo

stabilityai
/
sd-vae-ft-mse-original

Text-to-Image
stable-diffusion
stable-diffusion-diffusers
Model card Files Community
14
multimodalart HF staff commited on
Commit
d17ff5f
•
1 Parent(s): 833c4f1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -20
README.md CHANGED
@@ -9,7 +9,8 @@ inference: false
9
  # Improved Autoencoders
10
 
11
  ## Utilizing
12
- These weights are intended to be used with the original [CompVis Stable Diffusion codebase](https://github.com/CompVis/stable-diffusion). If you are looking for the model to use with the D🧨iffusers library, [come here](https://huggingface.co/CompVis/stabilityai/sd-vae-ft-mse).
 
13
  ## Decoder Finetuning
14
  We publish two kl-f8 autoencoder versions, finetuned from the original [kl-f8 autoencoder](https://github.com/CompVis/latent-diffusion#pretrained-autoencoding-models).
15
  The first, _ft-EMA_, was resumed from the original checkpoint, trained for 313198 steps and uses EMA weights.
@@ -44,41 +45,29 @@ _Visualization of reconstructions on 256x256 images from the COCO2017 validatio
44
  <p align="center">
45
  <br>
46
  <b>
47
- 256x256: ft-MSE (top), ft-EMA (middle), original (right)</b>
48
  </p>
49
 
50
  <p align="center">
51
- <img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00025.png" />
52
- <img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00025.png" />
53
- <img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00025.png" />
54
  </p>
55
 
56
  <p align="center">
57
- <img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00011.png" />
58
- <img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00011.png" />
59
- <img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00011.png" />
60
  </p>
61
 
62
  <p align="center">
63
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00037.png />
64
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00037.png />
65
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00037.png />
66
  </p>
67
 
68
  <p align="center">
69
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00043.png />
70
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00043.png />
71
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00043.png />
72
  </p>
73
 
74
  <p align="center">
75
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00053.png />
76
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00053.png />
77
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00053.png />
78
  </p>
79
 
80
  <p align="center">
81
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00029.png />
82
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00029.png />
83
- <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00029.png />
84
  </p>
 
9
  # Improved Autoencoders
10
 
11
  ## Utilizing
12
+ These weights are intended to be used with the original [CompVis Stable Diffusion codebase](https://github.com/CompVis/stable-diffusion). If you are looking for the model to use with the 🧨 diffusers library, [come here](https://huggingface.co/CompVis/stabilityai/sd-vae-ft-ema).
13
+
14
  ## Decoder Finetuning
15
  We publish two kl-f8 autoencoder versions, finetuned from the original [kl-f8 autoencoder](https://github.com/CompVis/latent-diffusion#pretrained-autoencoding-models).
16
  The first, _ft-EMA_, was resumed from the original checkpoint, trained for 313198 steps and uses EMA weights.
 
45
  <p align="center">
46
  <br>
47
  <b>
48
+ 256x256: ft-EMA (left), ft-MSE (middle), original (right)</b>
49
  </p>
50
 
51
  <p align="center">
52
+ <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00025_merged.png />
 
 
53
  </p>
54
 
55
  <p align="center">
56
+ <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00011_merged.png />
 
 
57
  </p>
58
 
59
  <p align="center">
60
+ <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00037_merged.png />
 
 
61
  </p>
62
 
63
  <p align="center">
64
+ <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00043_merged.png />
 
 
65
  </p>
66
 
67
  <p align="center">
68
+ <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00053_merged.png />
 
 
69
  </p>
70
 
71
  <p align="center">
72
+ <img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00029_merged.png />
 
 
73
  </p>