---
base_model: runwayml/stable-diffusion-v1-5
license: creativeml-openrail-m
language:
- en
library_name: diffusers
pipeline_tag: text-to-image
inference: true
---

[Original model page](https://civitai.com/models/25694/epicrealism)

<details>
  <summary>Examples</summary>

|                                 |                                 |
|---------------------------------|---------------------------------|
| ![](.hf/readme/examples/1.jpeg) | ![](.hf/readme/examples/2.jpeg) |
| ![](.hf/readme/examples/3.jpeg) | ![](.hf/readme/examples/4.jpeg) |
| ![](.hf/readme/examples/5.jpeg) | ![](.hf/readme/examples/6.jpeg) |
| ![](.hf/readme/examples/7.jpeg) | ![](.hf/readme/examples/8.jpeg) |

-----------

</details>

## Natural Sin Final and last of epiCRealism

~~Since SDXL is right around the corner~~, let's say it is the final version for now since I put a lot effort into it
and probably cannot do much more.

I tried to refine the understanding of the Prompts, Hands and of course the Realism.\
Let's see what you guys can do with it.

Advices:

- use simple prompts
- no need to use keywords like "masterpiece, best quality, 8k, intricate, high detail" or "(extremely detailed face), (
	extremely detailed hands), (extremely detailed hair)" since it doesn't produce appreciable change
- use simple negatives or small negative embeddings. gives most realistic look (check samples to get an idea of
	negatives i used)
- add "asian, chinese" to negative if you're looking for ethnicities other than Asian
- Light, shadows, and details are excellent without extra keywords
- If you're looking for a natural effect, avoid "cinematic"
- avoid using "1girl" since it pushes things to render/anime style
- to much description of the face will turn out bad mostly
- no extra noise-offset needed, but u can if you like to 😉

## How to use?

Prompt: simple explanation of the image (try first without extra keywords)\
Negative: "cartoon, painting, illustration, (worst quality, low quality, normal quality:2)"\
Steps: >50 (if image has errors or artefacts use higher steps)\
CFG Scale: 5.0 (higher config scale can lose realism, depends on prompt, sampler and steps)\
Size: 512x768 or 768x512

## Use it with 🧨 diffusers

This model can be used just like any other Stable Diffusion model. For more information,
please have a look at the [Diffusers](https://huggingface.co/docs/diffusers/index).

```python
import torch
from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained(
  'kamillaova/epiCRealism',
  torch_dtype=torch.float16
).to('cuda')

img = pipe(
  prompt='positive prompt',
  negative_prompt='negative prompt',

  guidance_scale=5.0,
  num_inference_steps=50,
  width=512, height=768,
).images[0]

img.save('image.png')
```