Downloads · 30 days
16
12% of all-time downloads
Bshiblobble/Flawed_CoMIX
Flawed_CoMIX is a machine learning model from Bshiblobble. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for diffusers. The card lists the license as openrail.
Downloads · 30 days
16
12% of all-time downloads
All-time downloads
137
Public
Repo size
7 GB
Likes
1
Public
Click a slice to open those files.
.safetensors6.9 GB · 99%
From the Hugging Face model README
🟡Flawed CoMix🟡
A checkpoint based on the NoobEQ finetune by Anzhc, merged with a combination of GloRAs I trained with the intent of getting a "western comicbook" look. It didn't turn out as I expected but can still generate really good images, it's just probably the least beginner-friendly distant offspring of SDXL. If you're looking for a checkpoint and have virtually no experience or just copy prompts from different websites, this checkpoint is not for you, you might have better chances with my SliMix
🟡VAE:
This checkpoint uses the "MS-LC-EQ-D-VR VAE B3" from Anzhc's repository and WILL produce blurry messes without it, download it here I did not test the other BX VAEs. I believe anything from B4 to B7 should still work at least decently with it.
🟡Basic use:
masterpiece, best quality at the start of your prompt.worst quality, low quality, lowres, bad anatomy, bad hands, jpeg artifacts, border, watermark in negative, add more negative as needed.🟡Advanced use:
You can add any of the following to your prompt.
painterly.lineart, ligne claire.high contrast to reinforce the contrast between dark values and light values.<color> theme to give the image a distinct color theme.color theme 1, color theme 2, color contrast to create an image with two contrasty color themes (might require weight tinkering because some are stronger than others).You can always look at the preview images to get ideas of what is possible and in fact I feel like this model can output a much wider style of images.
🟡Recommended settings:
🟡Others:
Forge Couple extension assuming your version of Forge or reForge supports ithalftone in negative can help with that.🟡Previews:
<style> .custom-table td { width: 16.161616161616%; } .custom-image-container { position: relative; width: 100%; height: 50%; border-radius: 0.5em; overflow: hidden; align-items: center; } .custom-image { width: auto; height: 50%; border-radius: 0.5em; object-fit: cover; transition: all 0.3s ease; } .custom-image-container:hover .custom-image { transform: scale(1.2); } </style> <table class="custom-table"> <tr> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/01.png" alt="Preview"> </div> </td> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/02.png" alt="Preview"> </div> </td> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/03.png" alt="Preview"> </div> </td> </tr> <tr> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/04.png" alt="Preview"> </div> </td> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/05.png" alt="Preview"> </div> </td> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/06.png" alt="Preview"> </div> </td> </tr> <tr> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/07.png" alt="Preview"> </div> </td> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/08.png" alt="Preview"> </div> </td> <td> <div class="custom-image-container"> <img class="custom-image" src="previews/09.png" alt="Preview"> </div> </td> </tr> </table>🟡Easy Download: