---
license: other
datasets:
- p1atdev/pvc
language:
- en
library_name: diffusers
tags:
- text-to-image
- stable-diffusion
- safetensors
widget:
- text: masterpiece, best quality, exceptional, 1girl, bangs, bare shoulders, beret, black hair, black shorts, blue hair, bracelet, breasts, buttons, colored inner hair, double-breasted, eyewear removed, green headwear, green jacket, grey eyes, grey sky, hat, jacket, jewelry, long hair, looking at viewer, multicolored hair, neck ring, o-ring, off shoulder, rain, round eyewear, shorts, sidelocks, small breasts, solo, sunglasses, wavy hair, wet, zipper
  example_title: The WD1.5 girl
- text: masterpiece, best quality, exceptional, 1girl, blonde hair, hat, baseball cap, aqua eyes, earrings, hoop earrings, yellow shirt, looking at viewer, upper body, simple background
  example_title: The blonde hair girl
- text: masterpiece, best quality, exceptional, 1girl, cat ears, red hair, long hair, hairpin, swept bangs, yellow eyes, black jacket, white shirt, blue tie, white gloves, hand up, upper body, looking at viewer, buildings
  example_title: A red hair girl
- text: nendoroid, masterpiece, best quality, exceptional, 1girl, cat ears, red hair, long hair, hairpin, swept bangs, yellow eyes, black jacket, white shirt, blue tie, white gloves, hand up, upper body, looking at viewer, 
  example_title: nendoroid style
- text: figma, masterpiece, best quality, exceptional, 1girl, cat ears, red hair, long hair, hairpin, swept bangs, yellow eyes, black jacket, white shirt, blue tie, white gloves, hand up, upper body, looking at viewer, buildings
  example_title: figma style
---

# PVC v3

This model is a latent diffusion model finetuned on Waifu Diffusion v1.5 beta 2 with PVC figure images. 
You can use Danbooru tags to generate images.

## Downloads

<div class="flex flex-col dark:bg-gray-900 rounded-md divide-y dark:divide-gray-800">

<div class="flex justify-between px-4 py-2">
  <a class="underline" href="https://huggingface.co/p1atdev/pvc-v3/resolve/checkpoints/pvc-v3-fp16.safetensors">pvc-v3-fp16.safetensors</a>
  <div>2.58 GB</div>
</div>

<div class="flex justify-between px-4 py-2">
  <a class="underline" href="https://huggingface.co/p1atdev/pvc-v3/resolve/checkpoints/pvc-v3-fp16.ckpt">pvc-v3-fp16.ckpt</a>
  <div>2.58 GB</div>
</div>

<div class="flex justify-between px-4 py-2">
  <a class="underline" href="https://huggingface.co/p1atdev/pvc-v3/resolve/checkpoints/pvc-v3-fp32.safetensors">pvc-v3-fp32.safetensors</a>
  <div>5.16 GB</div>
</div>

<div class="flex justify-between px-4 py-2">
  <a class="underline" href="https://huggingface.co/p1atdev/pvc-v3/resolve/checkpoints/pvc-v3-fp32.ckpt">pvc-v3-fp32.ckpt</a>
  <div>5.16 GB</div>
</div>

<div class="flex justify-between px-4 py-1">
  <a class="underline opacity-75" href="https://huggingface.co/p1atdev/pvc-v3/tree/checkpoints">Show all</a>
</div>

</div>

## Prompt guide

### Trigger words

- `pvc` means the pvc material style but not needed always.

- `figma` is the figure style that has joints, and more tend to be product thumbnail images. Use with `doll joints` to get better joints.

- `nendoroid` means the style of chibi figures. Use with `chibi` to get better results.

### Tips

The PVC figure style is closer to the anime style than to the realistic style. 
So, it is recommended to put `anime` to **positive** prompt and `realistic` to **negative** prompt to get better results sometimes.
If you want to avoid too realistic faces, try this! Setting `realistic` to negative does not affect the pvc style, only the faces of the characters. 



## Examples




## Training information

<details>
<table>
<thead>
<tr><th>Parameter</td><td>Value</th></tr>
</thead>
<tbody>
  
<tr><td>Service</td><td>Runpod</td></tr>
<tr><td>GPU</td><td>A5000</td></tr>
<tr><td>Notebook</td><td><a href="https://github.com/Linaqruf/kohya-trainer/blob/main/kohya-trainer.ipynb" target="_blank">Linaqruf/kohya-trainer</a></td></tr>
<tr><td>Cost</td><td>about $2</td></tr>
<tr><td>Hours</td><td>about 6 hours</td></tr>

<tr><td>Dataset</td><td>7467 images from p1atdev/pvc</td></tr>
<tr><td>Resolution</td><td>896</td></tr>
<tr><td>Epochs</td><td>5</td></tr>
<tr><td>Optimizer</td><td>Lion</td></tr>
<tr><td>LR</td><td>4e-7</td></tr>
<tr><td>Scheduler</td><td>cosine_with_restarts</td></tr>
<tr><td>Train Batch Size</td><td>1</td></tr>

</tbody>
</table>
</details>


## License
PVC v3 is released under the Fair AI Public License 1.0-SD (https://freedevproject.org/faipl-1.0-sd/). If any derivative of this model is made, please share your changes accordingly. Special thanks to ronsor/undeleted (https://undeleted.ronsor.com/) for help with the license.