---
license: creativeml-openrail-m
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
inference: false
---
# Improved Autoencoders
## Decoder Finetuning
We publish two kl-f8 autoencoder versions, finetuned from the original [kl-f8 autoencoder](https://github.com/CompVis/latent-diffusion#pretrained-autoencoding-models).
The first, _ft-EMA_, was resumed from the original checkpoint, trained for 313198 steps and uses EMA weights.
The second, _ft-MSE_, was resumed from _ft-EMA_ and uses EMA weights and was trained for another 280k steps using a re-weighted loss, with more emphasis 
on MSE reconstruction (producing somewhat ``smoother'' outputs).
To keep compatibility with existing models, only the decoder part was finetuned; the checkpoints can be used as a drop-in replacement for the existing autoencoder.

_Original kl-f8 VAE vs f8-ft-EMA vs f8-ft-MSE_

## Evaluation 
### COCO 2017 (256x256, val, 5000 images)
| Model    | train steps | rFID | PSNR         | SSIM          | PSIM          | Link                                                                              | Comments                                                                                        
|----------|---------|------|--------------|---------------|---------------|-----------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------|
|          |         |      |              |               |               |                                                                                   |                                                                                                 |
| original | 246803        | 4.99 | 23.4 +/- 3.8 | 0.69 +/- 0.14 | 1.01 +/- 0.28 | https://ommer-lab.com/files/latent-diffusion/kl-f8.zip                            | as used in SD                                                                                   |
| ft-EMA   | 560001        | 4.42 | 23.8 +/- 3.9 | 0.69 +/- 0.13 | 0.96 +/- 0.27 | https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/blob/main/ckpt_00560000-kat-ema-pruned.ckpt | slightly better overall, with EMA                                                               |
| ft-MSE   | 840001        | 4.70 | 24.5 +/- 3.7 | 0.71 +/- 0.13 | 0.92 +/- 0.27 | https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/blob/main/ckpt_00840000-kat-ema-pruned.ckpt | resumed with EMA from ft-EMA, emphasis on MSE (rec. loss = MSE + 0.1 * LPIPS), smoother outputs |


### LAION-Aesthetics 5+ (256x256, subset, 10000 images)
| Model    | train steps | rFID | PSNR         | SSIM          | PSIM          | Link                                                                              | Comments                                                                                        
|----------|-----------|------|--------------|---------------|---------------|-----------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------|
|          |           |      |              |               |               |                                                                                   |                                                                                                 |
| original | 246803         | 2.61 | 26.0 +/- 4.4 | 0.81 +/- 0.12 | 0.75 +/- 0.36 | https://ommer-lab.com/files/latent-diffusion/kl-f8.zip                            | as used in SD                                                                                   |
| ft-EMA   | 560001          | 1.77 | 26.7 +/- 4.8 | 0.82 +/- 0.12 | 0.67 +/- 0.34 | https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/blob/main/ckpt_00560000-kat-ema-pruned.ckpt | slightly better overall, with EMA                                                               |
| ft-MSE   | 840001          | 1.88 | 27.3 +/- 4.7 | 0.83 +/- 0.11 | 0.65 +/- 0.34 | https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/blob/main/ckpt_00840000-kat-ema-pruned.ckpt | resumed with EMA from ft-EMA, emphasis on MSE (rec. loss = MSE + 0.1 * LPIPS), smoother outputs |


### Visual
_Visualization of reconstructions on  256x256 images from the COCO2017 validation dataset._ 

<p align="center">
  <br>
  <b>
256x256: ft-MSE (top), ft-EMA (middle), original (right)</b>
</p>

<p align="center">
<img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00025.png" />
<img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00025.png" />
<img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00025.png" />
</p>

<p align="center">
<img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00011.png" />
<img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00011.png" />
<img src="https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00011.png" />
</p>

<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00037.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00037.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00037.png />
</p>

<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00043.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00043.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00043.png />
</p>

<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00053.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00053.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00053.png />
</p>

<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-mse/00029.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/ft-ema/00029.png />
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/original/00029.png />
</p>