---
license: openrail++
language:
- en
library_name: diffusers
pipeline_tag: text-to-image
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: p1x3l16
tags:
- lora
- stable-diffusion-xl
- text-to-image
- diffusers
- style
- pixel-art
- pixelart
- 16-bit
- retro
- game-art
- snes
- 8-bit
- retro-style
- chunky-pixels
- snes-era
- limited-palette
- clean-dithering
- nostalgic
- vintage
widget:
  - text: "p1x3l16, a wizard's tower on a cliff"
  - text: "p1x3l16, a treasure chest in a cave"
  - text: "p1x3l16, a dragon guarding a castle"
---

# 16-bit Pixel Art SDXL LoRA for Retro Game-Art

## What it does
Retro 16-bit pixel art in the style of classic SNES-era games, featuring a crisp limited palette, clean dithering, chunky readable pixels, and nostalgic game-art composition.

## How to use
Trigger: p1x3l16. Suggested strength: 0.85. Recommended sampler: Euler a or Euler a with noise reducer.

## Best for
- Wizard's towers on cliffs
- Treasure chests in caves
- Dragons guarding castles
- Cozy tavern interiors
- Heroes walking through deserts

## Tips
Adjust the strength to fine-tune the pixel art style. Use higher strengths for more pronounced retro game aesthetics.
## Example Prompts

```
p1x3l16 cyberpunk night cityscape with neon lights glowing in a futuristic urban scene
```

```
p1x3l16 hero warrior wielding sword and shield in a fantasy RPG game setting under moonlight
```

```
p1x3l16 cozy living room interior with fireplace and bookshelf, warm lighting for a nostalgic home
```

```
p1x3l16 space explorer landing on alien planet with strange flora, stars visible through the sky
```

```
p1x3l16 character portrait of an adventurer in leather armor, looking determined at dusk setting
```

```
p1x3l16 abstract geometric shapes and patterns in vibrant colors for a digital art poster design
```

## Trigger word

Activate with `p1x3l16`. Recommended strength: **0.9**.

## Showcase

![showcase 1](./showcase/showcase_01.png)
![showcase 2](./showcase/showcase_02.png)
![showcase 3](./showcase/showcase_03.png)
![showcase 4](./showcase/showcase_04.png)
![showcase 5](./showcase/showcase_05.png)
![showcase 6](./showcase/showcase_06.png)
![showcase 7](./showcase/showcase_07.png)
![showcase 8](./showcase/showcase_08.png)

## Usage with `diffusers`

```python
from diffusers import AutoPipelineForText2Image
import torch

pipe = AutoPipelineForText2Image.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16).to("cuda")
pipe.load_lora_weights("strkyyy/pixel-art-16bit")
image = pipe("p1x3l16, your prompt", num_inference_steps=30).images[0]
```