|
--- |
|
license: creativeml-openrail-m |
|
language: |
|
- en |
|
tags: |
|
- stable-diffusion |
|
- stable-diffusion-diffusers |
|
- text-to-image |
|
- safetensors |
|
- diffusers |
|
- artwork |
|
- HDR photography |
|
- safetensors |
|
- photos |
|
inference: true |
|
--- |
|
|
|
# Foto Assisted Diffusion (FAD)_V0 |
|
|
|
This model is meant to mimic a modern HDR photography style |
|
|
|
It was trained on 600 HDR images on SD1.5 and works best at **768x768** resolutions |
|
|
|
Merged with one of my own models for illustrations and drawings, to increase flexibility |
|
|
|
# Features: |
|
* **No additional licensing** |
|
* **Multi-resolution support** |
|
* **HDR photographic outputs** |
|
* **No Hi-Res fix required** |
|
* [**Spreadsheet with supported resolutions, keywords for prompting and other useful hints/tips**](https://docs.google.com/spreadsheets/d/1RGRLZhgiFtLMm5Pg8qK0YMc6wr6uvj9-XdiFM877Pp0/edit#gid=364842308) |
|
|
|
|
|
# Example Cards: |
|
|
|
Below you will find some example cards that this model is capable of outputting. |
|
You can acquire the images used here: [HF](https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/tree/main/Model%20Examples) or |
|
[Google Drive](https://docs.google.com/spreadsheets/d/1RGRLZhgiFtLMm5Pg8qK0YMc6wr6uvj9-XdiFM877Pp0/edit#gid=364842308). |
|
|
|
Google Drive gives you them all at once without needing to clone the repo, which is easier. |
|
|
|
If you decide to clone it, set ``` GIT_LFS_SKIP_SMUDGE=1 ``` to skip downloading large files |
|
|
|
Place them into an EXIF viewer such as the built in "PNG Info" tab in the popular Auto1111 repository to quickly copy the parameters and replicate them! |
|
## 768x768 Food |
|
<img src="https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/resolve/main/768x768%20Food.jpg" style="max-width: 800px;" width="100%"/> |
|
|
|
|
|
## 768x768 Landscapes |
|
<img src="https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/resolve/main/768x768%20Landscapes.jpg" style="max-width: 800px;" width="100%"/> |
|
|
|
## 768x768 People |
|
<img src="https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/resolve/main/768x768%20People.jpg" style="max-width: 800px;" width="100%"/> |
|
|
|
## 768x768 Random |
|
<img src="https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/resolve/main/768x768%20Random.jpg" style="max-width: 800px;" width="100%"/> |
|
|
|
## 512x512 Artwork |
|
<img src="https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/resolve/main/512x512%20Artwork.jpg" style="max-width: 800px;" width="100%"/> |
|
|
|
## 512x512 Photos |
|
<img src="https://huggingface.co/Dunkindont/Foto-Assisted-Diffusion-FAD_V0/resolve/main/512x512%20Photo.jpg" style="max-width: 800px;" width="100%"/> |
|
|
|
## Cloud Support |
|
Sinkin kindly hosted our model. [Click here to run it on the cloud](https://sinkin.ai/m/V6vYoaL)! |
|
|
|
## License |
|
*My motivation for making this model was to have a free, non-restricted model for the community to use and for startups.* |
|
|
|
*I was noticing the models people gravitated towards, were merged models which had prior license requirements from the people who trained them.* |
|
|
|
*This was just a fun project I put together for you guys.* |
|
|
|
*My fun ended when I posted the results :D* |
|
|
|
*Enjoy! Sharing is caring :)* |
