README.md

---
license: other
license_name: faipl-1.0-sd
license_link: https://freedevproject.org/faipl-1.0-sd/
language:
- en
tags:
  - text-to-image
  - stable-diffusion
  - safetensors
  - stable-diffusion-xl
---
<style>
  .title{
    font-size: 2.5em;
    letter-spacing: 0.01em;
    padding: 0.5em 0;
  }
</style>

# models
- [Zipang XL test03](#test03) (recommended)
- [Ponypang_XL_test](#potest)
- [Zipang XL test01](#test01)
---

<br />

<a id="test03"></a> 
<h1 class="title">
<span>Zipang XL test03</span>
</h1>
  -4000+ twitter images trained & 9000+ images merged model<br/>
  <br/>
 I don't like the look a bit, but the controls are much improved. <br/>
 <br/>

[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/zipang_XL_test03.fp16.safetensors?download=true)<br/>
[myJS LoRA (dedicated)](#myjs_lora)
[myJC LoRA (dedicated)](#myjc_lora)

<table>
  <tr>
    <td>
<a href="https://img92.pixhost.to/images/612/443376066_z03_01.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/612/443376066_z03_01.jpg" alt="sample1">
      </div>
		</a>
<a href="https://img92.pixhost.to/images/612/443376069_z03_02.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/612/443376069_z03_02.jpg" alt="sample2">
      </div>
		</a>
    </td>
    <td>
<a href="https://img92.pixhost.to/images/612/443376071_z03_03.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/612/443376071_z03_03.jpg" alt="sample3">
      </div>
		  </a>
<a href="https://img92.pixhost.to/images/612/443376074_z03_04.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/612/443376074_z03_04.jpg" alt="sample4">
    </td>
      </div>
		</a>
    <td>
<a href="https://img92.pixhost.to/images/612/443376076_z03_05.jpg" target=”_blank”>
<div>
        <img src="https://t92.pixhost.to/thumbs/612/443376076_z03_05.jpg" alt="sample5">
      </div>
		</a>
<a href="https://img92.pixhost.to/images/612/443376088_z03_06.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/612/443376088_z03_06.jpg" alt="sample6">
      </div>
		</a>
    </td>
  </tr>
</table>
-refer to pnginfo 
<br />

---

# - base model
-<a href="https://huggingface.co/cagliostrolab/animagine-xl-3.0">Animagine XL 3.0</a>
  -great XL model!<br />
-<a href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">stable-diffusion-xl-base-1.0 </a>
<br />
## - trigger

```bash
japanese
european
```
<br />

## - quality tags
```bash
masterpiece, best quality, realistic, 
```
<br />

## - negative tags
```bash
bad anatomy, worst quality, low quality
```
<br />

## - sampler
```bash
DPM++ 2M SDE Heun Karras :24-28 steps
(LCM)DPM++ 2S a Karras :6-8 steps
(Lightning)DPM++ 2M SGMUniform :8-10 steps
```
<br />

## - cfg scale
```bash
5-7 (LCM:3-5) (Lightning:1-1.3)
```
<br />

## - Resolution
```bash
768 x1024, 1024 x 768
1024 x 1024
896 x 1152, 1152 x 896
512 x 768, 768 x 512 (hires.fix required)
```
<br />

# sample prompt
[<img src=https://t92.pixhost.to/thumbs/612/443376095_z03_07.jpg />](https://img92.pixhost.to/images/612/443376095_z03_07.jpg)
```bash
masterpiece, best quality, natural lighting,
japanese, realistic, 1girl, solo, cute,
bangs, brown hair, ponytail, brown eyes, smile,
long white coat, zipper, snow fall,
v, looking at viewer, outdoors, photo background, upper body, front view
Negative prompt: bad anatomy, worst quality, low quality, nsfw
Steps: 28, Sampler: DPM++ 2M SDE Heun Karras, CFG scale: 3, Seed: 2206710877, Size: 896x1152, Model hash: 8a9978e263, Model: zipang_XL_test03.fp16, Version: v1.7.0
```

<br />

-[LCM](https://huggingface.co/latent-consistency/lcm-lora-sdxl)
[<img src=https://t92.pixhost.to/thumbs/612/443376096_z03_08.jpg />](https://img92.pixhost.to/images/612/443376096_z03_08.jpg)
```bash
masterpiece, best quality,
15yo, japanese, realistic, 1girl, solo, cute,
bangs, black hair, long hair, black eyes, smile,
cat ears, wristwatch, school uniform,
indoors, shop, looking at viewer
<lora:lcm-lora-sdxl:0.3>
Negative prompt: bad anatomy, worst quality, low quality
Steps: 8, Sampler: DPM++ 2S a Karras, CFG scale: 3, Seed: 246389822, Size: 896x1152, Model hash: 8a9978e263, Model: zipang_XL_test03.fp16, Lora hashes: "lcm-lora-sdxl: 2fa7e8e56b09", Version: v1.7.0
```
<br />

<br />

-[SDXL Lightning](https://huggingface.co/ByteDance/SDXL-Lightning) (too fast: recommended)
[<img src=https://t93.pixhost.to/thumbs/1137/448456359_20240224143137_zipang_xl_test03-fp16_1747615192.jpg />](https://img93.pixhost.to/images/1137/448456359_20240224143137_zipang_xl_test03-fp16_1747615192.jpg)
```bash
masterpiece, best quality, cute, japanese,
14yo, 1girl, bangs, black hair, twintails, police uniform, tokyo street, wind, posing, smile, upper body
<lora:sdxl_lightning_8step_lora:1>
Negative prompt: bad anatomy, worst quality, low quality
Steps: 8, Sampler: DPM++ 2M SGMUniform, CFG scale: 1.3, Seed: 1747615192, Size: 896x1152, Model hash: 8a9978e263, Model: zipang_XL_test03.fp16, Version: f0.0.15v1.8.0rc-latest-229-g54c89503
```
<br />
<br />
-merge & train<br />

```bash
base1:merge Animagine XL and stable-diffusion-xl-base-1.0
      mbw(0.1,1,1,1,1,0,0,0.1,0.1,0,0.3,0,1,0.5,1,0.5,1,1,1,1)
base2:trained 4000+images:manual tagging (base model:base1)(Adafactor:lr:1e-04)
model:trained 9000+images:wd14 tagger(swin v2) (base model:base2)(Adafactor:lr:1e-04)
```
<br />
<br />

---

<a id="myjs_lora"></a> 

# my JS LoRA for test03

-natural face

[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/myjs_XL_zipang_test03.safetensors?download=true)<br/>
### trigger:
```bash
myjs
```
<br />
<table>
<tr>
<td bgcolor=#ddddff>original</td>
<td bgcolor=#ddddff>LoRA myjs</td>
</tr><tr>
<td>
      <a href="https://img93.pixhost.to/images/1208/449250745_myjs_zipang_01_org.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1208/449250745_myjs_zipang_01_org.jpg" alt="sample1">
      </div>
      </a>
</td>
<td>
      <a href="https://img93.pixhost.to/images/1208/449250736_myjs_zipang_01_lora.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1208/449250736_myjs_zipang_01_lora.jpg" alt="sample1">
      </div>
      </a>
</td>
</tr><tr>
<td>
      <a href="https://img93.pixhost.to/images/1208/449250751_myjs_zipang_02_org.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1208/449250751_myjs_zipang_02_org.jpg" alt="sample1">
      </div>
      </a>
</td>
<td>
      <a href="https://img93.pixhost.to/images/1208/449250750_myjs_zipang_02_lora.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1208/449250750_myjs_zipang_02_lora.jpg" alt="sample1">
      </div>
      </a>
</td>
</tr>
</table>

-refer to pnginfo 
<br />

---

<a id="myjc_lora"></a> 

# my JC LoRA for test03

-natural face

[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/myjc_v1_XL.safetensors?download=true)<br/>
### trigger:
```bash
myjc
```
<br />
<table>
<tr>
<td bgcolor=#ddddff>original</td>
<td bgcolor=#ddddff>LoRA myjc</td>
</tr><tr>
<td>
      <a href="https://img92.pixhost.to/images/634/443649241_01_orig.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/634/443649241_01_orig.jpg" alt="sample1">
      </div>
      </a>
</td>
<td>
      <a href="https://img92.pixhost.to/images/634/443649237_01_myjc.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/634/443649237_01_myjc.jpg" alt="sample1">
      </div>
      </a>
</td>
</tr><tr>
<td>
      <a href="https://img92.pixhost.to/images/634/443649246_02_orig.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/634/443649246_02_orig.jpg" alt="sample1">
      </div>
      </a>
</td>
<td>
      <a href="https://img92.pixhost.to/images/634/443649243_02_myjc.jpg" target=”_blank”>
      <div>
        <img src="https://t92.pixhost.to/thumbs/634/443649243_02_myjc.jpg" alt="sample1">
      </div>
      </a>
</td>
</tr>
</table>

-refer to pnginfo 
<br />

---

<a id="potest"></a> 
<h1 class="title">
  <span>Ponypang XL test</span>
</h1>
  -4000+ twitter images trained & 9000+ images merged model<br/>
  <br/>
-experimental<br/>
-many bankruptcies<br/>
-low quality<br/>
 <br/>

[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/ponypang_XL_test.safetensors?download=true)<br/>
[LoRA (dedicated)](#myjs_po_lora)<br/>

<table>
  <tr>
    <td>
<a href="https://img93.pixhost.to/images/1139/448474065_popang01.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1139/448474065_popang01.jpg" alt="sample1">
      </div>
		</a>
<a href="https://img93.pixhost.to/images/1139/448474070_popang02.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1139/448474070_popang02.jpg" alt="sample2">
      </div>
		</a>
    </td>
    <td>
<a href="https://img93.pixhost.to/images/1139/448476416_popang03.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1139/448476416_popang03.jpg" alt="sample3">
      </div>
		  </a>
<a href="https://img93.pixhost.to/images/1139/448478736_popang04.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1139/448478736_popang04.jpg" alt="sample4">
    </td>
      </div>
		</a>
    <td>
<a href="https://img93.pixhost.to/images/1139/448481436_popang05.jpg" target=”_blank”>
<div>
        <img src="https://t93.pixhost.to/thumbs/1139/448481436_popang05.jpg" alt="sample5">
      </div>
		</a>
<a href="https://img93.pixhost.to/images/1139/448482903_popang06.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1139/448482903_popang06.jpg" alt="sample6">
      </div>
		</a>
    </td>
  </tr>
</table>
-refer to pnginfo 

---

# - base model
-<a href="https://civitai.com/models/257749/pony-diffusion-v6-xl">Pony Diffusion V6 XL</a>
  -great XL model!<br />
-<a href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">stable-diffusion-xl-base-1.0 </a>
<br />
## - trigger

```bash
japanese
european
```
<br />

## - quality tags
```bash
score_9, score_8_up, score_7_up, best quality, masterpiece, 
```
<br />

## - negative tags
```bash
worst quality, low quality, normal quality, messy drawing, amateur drawing, lowres, bad anatomy, bad hands, source_furry, source_pony, source_cartoon, comic, source filmmaker, 3d
```
<br />

## - sampler
```bash
DPM++ 2M SDE Heun Karras, DPM++ 2S a Karras, UniPC :24-28 steps
(Lightning)DPM++ 2M SGMUniform :8-10 steps
```
<br />

## - cfg scale
```bash
3-5 (Lightning:1-1.3)
```
<br />

## - Resolution
```bash
768 x1024, 1024 x 768
1024 x 1024
896 x 1152, 1152 x 896
512 x 768, 768 x 512 (hires.fix required)
```
<br />

# sample prompt
[<img src=https://t93.pixhost.to/thumbs/1140/448485485_popang07.jpg />](https://img93.pixhost.to/images/1140/448485485_popang07.jpg)
```bash
score_9, score_8_up, score_7_up, best quality, masterpiece, cute, japanese,
15yo, 1girl, solo, smile, black hair, bun, blunt bangs, pink kimono, heart hands, japanese room, upper body
Negative prompt: worst quality, low quality, normal quality, messy drawing, amateur drawing, lowres, bad anatomy, bad hands, source_furry, source_pony, source_cartoon, comic, source filmmaker, 3d
Steps: 28, Sampler: DPM++ 2M SDE Heun Karras, CFG scale: 4, Seed: 4024721872, Size: 896x1152, Model hash: ee82fbfa99, Model: ponypang_XL_test, Version: f0.0.15v1.8.0rc-latest-229-g54c89503
```

<br />

-[SDXL Lightning](https://huggingface.co/ByteDance/SDXL-Lightning)
[<img src=https://t93.pixhost.to/thumbs/1140/448485490_popang08.jpg />](https://img93.pixhost.to/images/1140/448485490_popang08.jpg)
```bash
score_9, score_8_up, score_7_up, best quality, masterpiece, cute, japanese,
15yo, 1girl, solo, smile, black hair, bun, blunt bangs, pink kimono, heart hands, japanese room, upper body
<lora:sdxl_lightning_8step_lora:1>
Negative prompt: worst quality, low quality, normal quality, messy drawing, amateur drawing, lowres, bad anatomy, bad hands, source_furry, source_pony, source_cartoon, comic, source filmmaker, 3d
Steps: 8, Sampler: DPM++ 2M SGMUniform, CFG scale: 1.3, Seed: 1817917982, Size: 896x1152, Model hash: ee82fbfa99, Model: ponypang_XL_test, Version: f0.0.15v1.8.0rc-latest-229-g54c89503
```
<br />
<br />
-merge & train<br />

```bash
base1:merge Pony Diffusion V6 XL and stable-diffusion-xl-base-1.0
      mbw(1,1,0,1,1,1,0,1,1,0,0,0.5,1,1,1,1,1,0,1,0)
base2:trained 4000+images:manual tagging (base model:base1)(NAdam:lr:1e-04)
model:trained 9000+images:wd14 tagger(swin v2) (base model:base2)(NAdam:lr:1e-04)
```
<br />
<br />
---

<a id="myjs_po_lora"></a> 

# my JS LoRA for ponypang test

-natural face

[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/myjs_XL_ponypang.safetensors?download=true)<br/>
### trigger:
```bash
myjs
```
<br />
<table>
<tr>
<td bgcolor=#ddddff>original</td>
<td bgcolor=#ddddff>LoRA myjs</td>
</tr><tr>
<td>
      <a href="https://img93.pixhost.to/images/1160/448689070_myjs_po_none.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1160/448689070_myjs_po_none.jpg" alt="sample1">
      </div>
      </a>
</td>
<td>
      <a href="https://img92.pixhost.to/images/634/443649237_01_myjc.jpg" target=”_blank”>
      <div>
        <img src="https://t93.pixhost.to/thumbs/1160/448689068_myjs_po.jpg" alt="sample1">
      </div>
      </a>
</td>
</tr>
</table>
-refer to pnginfo 

<br />

---

<h1 class="title">
  <span>Zipang XL test02</span>
</h1>
  -failed<br/>
  
[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/zipang_XL_test02.fp16.safetensors?download=true)<br/>


---
<a id="test01"></a> 
<h1 class="title">
  <span>Zipang XL test01</span>
</h1>
  -9000+ twitter images trained & merged model<br/>

[Download](https://huggingface.co/deadman44/SDXL_Photoreal_Merged_Models/resolve/main/zipang_XL_test01.fp16.safetensors?download=true)<br/>

<table>
  <tr>
    <td>
<a href="https://img93.pixhost.to/images/255/438578262_zipangxl_sample01.jpg" target=”_blank”>
      <div>
        <img src="https://img93.pixhost.to/images/255/438578262_zipangxl_sample01.jpg" alt="sample1">
      </div>
		</a>
<a href="https://img93.pixhost.to/images/255/438578286_zipangxl_sample02.jpg" target=”_blank”>
      <div>
        <img src="https://img93.pixhost.to/images/255/438578286_zipangxl_sample02.jpg" alt="sample2">
      </div>
		</a>
    </td>
    <td>
<a href="https://img93.pixhost.to/images/255/438578311_zipangxl_sample03.jpg" target=”_blank”>
      <div>
        <img src="https://img93.pixhost.to/images/255/438578311_zipangxl_sample03.jpg" alt="sample3">
      </div>
		  </a>
<a href="https://img93.pixhost.to/images/255/438578339_zipangxl_sample04.jpg" target=”_blank”>
      <div>
        <img src="https://img93.pixhost.to/images/255/438578339_zipangxl_sample04.jpg" alt="sample4">
    </td>
      </div>
		</a>
    <td>
<a href="https://img93.pixhost.to/images/255/438584646_zipangxl_sample05_lcm.jpg" target=”_blank”>
<div>
        <img src="https://img93.pixhost.to/images/255/438584646_zipangxl_sample05_lcm.jpg" alt="sample5">
      </div>
		</a>
<a href="https://img93.pixhost.to/images/255/438584655_zipangxl_sample06_lcm.jpg" target=”_blank”>
      <div>
        <img src="https://img93.pixhost.to/images/255/438584655_zipangxl_sample06_lcm.jpg" alt="sample6">
      </div>
		</a>
    </td>
  </tr>
</table>
-refer to pnginfo 
<br />

---

# - base model
   <a href="https://huggingface.co/cagliostrolab/animagine-xl-3.0">Animagine XL 3.0</a>
  -great XL model!
<br />
<br />
## - trigger

```bash
japanese
european
```
<br />

## - quality tags
```bash
masterpiece, best quality, realistic, 
```
<br />

## - negative tags
```bash
bad anatomy, worst quality, low quality
```
<br />

## - sampler
```bash
DPM++ 2M SDE Heun Karras :24-28 steps
(LCM)DPM++ 2S a Karras :6-8 steps
```
<br />

## - cfg scale
```bash
5-7 (LCM:3-5)
```
<br />

# sample prompt
[<img src=https://img93.pixhost.to/images/256/438592852_zipangxl_sample07.jpg  style="max-width: 128px;" width="50%" />](https://img93.pixhost.to/images/256/438592852_zipangxl_sample07.jpg)
```bash
masterpiece, best quality, realistic,
natural lighting, day,
japanese, 1girl, solo, black hair, bangs, hair ornament, smile, parted lips, pink kimono, flower print, standing, looking at viewer, street
Negative prompt: bad anatomy, worst quality, low quality
Steps: 24, Sampler: DPM++ 2M SDE Heun Karras, CFG scale: 5, Seed: 3335242054, Size: 768x1024, Model hash: bb1f402b71, Model: zipang_XL_test01.fp16, Version: v1.7.0
```

<br />

-LCM
[<img src=https://img93.pixhost.to/images/256/438590779_zipangxl_sample08_lcm.jpg  style="max-width: 128px;" width="50%" />](https://img93.pixhost.to/images/256/438590779_zipangxl_sample08_lcm.jpg)
```bash
masterpiece, best quality, realistic,
natural lighting,
japanese, solo, 1girl, upper body, black hair, blunt bangs, long hair, black eye, grin, lips, long sleeves, looking at viewers, simple background
<lora:lcm-lora-sdxl:0.3>
Negative prompt: bad anatomy, worst quality, low quality
Steps: 6, Sampler: DPM++ 2S a Karras, CFG scale: 3, Seed: 282342251, Size: 768x1024, Model hash: bb1f402b71, Model: zipang_XL_test01.fp16, Lora hashes: "lcm-lora-sdxl: 2fa7e8e56b09", Version: v1.7.0
```
<br />
<br />
-merge & train<br />

```bash
model:trained 9000+images (base model:Animagine XL)(Adafactor:lr:1e-04)
```