<p align='center'> <img src="https://huggingface.co/Hosioka/Baka-Diffusion/resolve/main/Images/HeaderCurrent.png" align='center'> </p>

<h1 align='center'><b>·</b> Baka-Diffusion <b>·</b></h1>

<h3 align='center'> <a href='https://www.figma.com/file/1JYEljsTwm6qRwR665yI7w/Merging-lab%E3%80%8CHosioka-Fork%E3%80%8D?type=design&node-id=1-69&t=MPKBSTP9UWxp91vw-0'><b>·</b> MBW Resources <b>·</b></a> </h3>

<p align='center'><b>Baka-Diffusion</b> is a series of finetuned / U-Net Block merges made to push the limits of SD1.5 based models. Our models uses the Danbooru Tagging system. </p>

⚠️ Disclaimer

You are responsible for the content you generate, whether it is NSFW or SFW. The AI models do not contain explicit visual content that can be accessed easily.

⚒️ Model synopsis

Due to the nature of U-Net Block merging, some models function differently in CFG Scales. It is generally preferred to use the models in this repository with CFG 3-9. above this range are usable, but you may encounter artifacting, visible leftover noise, or color burn.

• Baka-Diffusion[S3D]
  • Baka-DiffusionV4[S3D] was created to achieve a soft 3D effect and replicate realistic lighting. It differs greatly from the regular Baka-Diffusion models, where the lighting has a very anime style.
  • This particular model is suited for higher resolutions. Instead of 512x768, use 600x896.
  • As of now this is the only model released.
  • The model is highly compatible with low rank networks such as LoRA / LyCORIS models.

<p align='center'> <img src='https://huggingface.co/Hosioka/Baka-Diffusion/resolve/main/Images/Baka-DiffusionS3D%20showcase.png'> <img src='https://huggingface.co/Hosioka/Baka-Diffusion/resolve/main/Images/LoRALycoris%20Preview%5BS3D%5D.png'>

To replicate my results and use the model as intended, please use the following default negative: (worst quality, low quality:1.2), lowres, bad anatomy, bad hands, nsfw. This negative served as a base when merging the model. Please avoid negative textual inversions, as they are designed for 2D anime models and using them on the [S3D] model will result in a style offset. </p>

🔧 Recommended Settings

• Default negative: (worst quality, low quality:1.2), lowres, bad anatomy, bad hands, nsfw
• CFG 3-9
• Clip skip 2 (Required!) <br></br>

📝 Note! <br></br> All the other model's promises have been cancelled. I've hit my limit and I'm just burnt out from trying to create anime models. Pretty soon, you'll find every Test Version of Baka-Diffusion models uploaded here. It's just... sigh... it's all too much. <img src='https://huggingface.co/Hosioka/Baka-Diffusion/resolve/main/Images/Screenshot%202023-07-30%20190021.png'> At this stage even though SDXL is brandnew with no proper foundations set in stone for anime. I'd like to think the SD1.5 architecture is outdated and is a waste of computing power. This was also the decicidng factor to stop SD1.5 and making models all together

📖 License

This repository is released under the CC BY-NC 2.0 license. The license specifies the following terms:

1. You are free to copy, distribute, display and perform the work, and to make derivative works, under the following conditions:
2. You must give the original author credit.
3. You may not use the work for commercial purposes.
4. You may Adapt: remix, transform, and build upon the material.
5. For any reuse or distribution, you must make clear to others the license terms of this work. <br></br>

🔗 Kudos!

• Erasing : https://github.com/rohitgandikota/erasing
• Runtime Block Merge: https://github.com/ashen-sensored/sd-webui-runtime-block-merge
• Super Merger: https://github.com/hako-mikan/sd-webui-supermerger
• KL-F8 Anime2 VAE: https://huggingface.co/hakurei/waifu-diffusion-v1-4/blob/main/vae/kl-f8-anime2.ckpt

<br></br> <p align='center'> <img src='https://huggingface.co/Hosioka/Baka-Diffusion/resolve/main/Images/S3D%20Eyes.png'> </p>