---
license: other
---
# Detailedprojectv4
Detailedprojectv4 is the fourth edition of detailedproject, also known as dpep. It's finetuned/natively trained on a varied dataset of detailed anime-styled artwork and scenery. Use `detailed background` to increase the amount of detail in backgrounds, as the most detailed artworks in the dataset were tagged with that.

# sleepymix3
Sleepymix3 is a mix using detailedprojectv4 as the base, with pieces mixed in. It's stylized and has high detail.

## Examples for sleepymix3

####
**1:**
![1](https://huggingface.co/closertodeath/detailedproject/resolve/main/mixes/example1.png))

**2:**
![2](https://huggingface.co/closertodeath/detailedproject/resolve/main/mixes/example4.png)

**3:**
![3](https://huggingface.co/closertodeath/detailedproject/resolve/main/mixes/example3.png)

### Prompt and settings for Example Images

**1:**
```
detailed background, masterpiece, best quality, 1girl, solo, (mystia lorelei, touhou, pink hair, purple eyes, wings, short hair, hat, ribbon, dress, frills:1.2), looking to the side, wavy mouth, outdoors, steam, rocks, onsen, steam, hot springs, water, partially submerged, flexing
Negative prompt: simple4, (bad child's drawing of scenery:1.6), (worst quality, low quality, normal quality, bad quality:1.7), (blurry, blurry background:1.5), depth of field, (simple background:1.4), flat color, flat shading, easynegative, bad-image, ((3d))
Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 20, Seed: 561393467, Size: 2432x1024, Model hash: 2fb7d63ed0, Model: sleepymix3, Denoising strength: 0.5, Clip skip: 2, ENSD: 31337, Mask blur: 4, MultiDiffusion upscaler: 4x-AnimeSharp, MultiDiffusion scale factor: 2, MultiDiffusion tile width: 112, MultiDiffusion tile height: 112, MultiDiffusion overlap: 32
```

**2:**
```
detailed background, masterpiece, best quality, 1girl, solo, (koakuma, touhou, red hair, red eyes, demon wings, hair ribbon, dress, detached sleeves, thigh-highs:1.2), looking away, angry, new york, new york cityscape, outdoors, nyc, curtsey
Negative prompt: simple4, (bad child's drawing of scenery:1.6), (worst quality, low quality, normal quality, bad quality:1.7), (blurry, blurry background:1.5), depth of field, (simple background:1.4), flat color, flat shading, easynegative, bad-image, ((3d))
Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 20, Seed: 896529780, Size: 512x512, Model hash: 2fb7d63ed0, Model: sleepymix3, Denoising strength: 0.5, Clip skip: 2, ENSD: 31337, Mask blur: 4
```

**3:**
```
detailed background, masterpiece, best quality, 1girl, solo, saigyouji yuyuko, touhou, pink hair, red eyes, hat, kimono, hair ornaments, rocks
Negative prompt: simple4, (bad child's drawing of scenery:1.6), (worst quality, low quality, normal quality, bad quality:1.7), (blurry, blurry background:1.5), depth of field, (simple background:1.4), flat color, flat shading, easynegative, bad-image, ((3d))
Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 20, Seed: 2866277638, Size: 512x512, Model hash: 2fb7d63ed0, Model: sleepymix3, Denoising strength: 0.5, Clip skip: 2, ENSD: 31337, Mask blur: 4
```

▼**Merging Process:**

## Merging Process Step 1:
| Model: A     | Model: B | Weight                                                                | Base alpha | Merge Name          |
| ------------ | -------- | --------------------------------------------------------------------- | ---------- | ------------------- |
| detailedpjv4 | BasilMix | 1,0.9,0.7,0.5,0.3,0.1,1,1,1,1,1,1,0,0,0,0,0,0,0,0.1,0.3,0.5,0.7,0.9,1 | 0          |     sleepymix1      |

## Step 2:
Now we merge this lora into sleepymix1 to create sleepymix2.

[beautiful detailed eyes lora](https://civitai.com/models/5693/beautiful-detailed-eyes) at 0.5

## Step 3:
Using sleepymix2, merge these two locons in to create sleepymix3.

[loraslayer2](https://huggingface.co/closertodeath/ctdlora/blob/main/loraslayer2.safetensors) at 0.4

[mem-locon](https://huggingface.co/closertodeath/ctdlora/blob/main/locon/mem-locon.safetensors) at 0.4


## License

You are free to:
1. Share — copy and redistribute the material in any medium or format
2. Adapt — remix, transform, and build upon the material, as long as you freely share the changes
   
Under the following terms: 
1. You cannot use the model to deliberately produce nor share illegal or harmful outputs or content
2. Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
3. You may not use the material for commercial purposes, whether it be as a service, sold as is or merged into other material.
4. If you grant access to a modified version of the model available to users over a network, you must make your modified model available to those users immediately.