Pixel-Party-XL Open-Source Text-to-Image Model - Freely Create Stunning Pixel Art Effects

Pixel Party Xl

Developed by pixelparty

A fully trained model based on SDXL, optimized for pixel art effects in text-to-image generation

Image Generation Open Source License:Openrail #Pixel Art Generation #Game Asset Creation #SDXL Fine-tuning

Downloads 87

Release Time : 8/8/2023

Model Overview

This model specializes in generating high-quality pixel art images, particularly suitable for asset creation in indie game development.

Model Features

Pixel Art Optimization

Specialized optimization for pixel art generation through full model training

Game Development Tool Integration

Supports common game development functions like map tile generation and animation production

Small-Size Optimization

Performs best on 128x128 canvas size, ideal for generating small items/characters

Model Capabilities

Pixel Art Generation

Game Asset Creation

Image Restoration

Character Redesign

Animation Interpolation

Use Cases

Game Development

Map Tile Generation

Generate pixel-style tiles required for game maps

Assets ready for direct use in game engines

Character Animation Production

Generate character movement and attack animations

Smooth pixel-style animation sequences

Digital Art

Pixel Art Creation

Generate various pixel-style artworks based on text descriptions

High-quality pixel-style images

🚀 Pixel Party XL

This is a full - scale model trained on SDXL to better adhere to pixel art styles. It offers a range of tools for indie game development and is available for personal project use.

🚀 Quick Start

This model is based on SDXL and trained to better adhere to pixel art styles. You can use it for your own projects, but hosting the model is prohibited.

examples

✨ Features

We are developing tools for indie game development, and currently have the following tools available:

Map tiles
Movement animations
Attack animations
Inpainting
Character reshaping
Animation interpolation

More features are in the pipeline! :D

If you'd like to support us or explore our other pixel art models, you can find us at PixelLab or on Discord.

📦 Installation

No specific installation steps are provided in the original README.

💻 Usage Examples

Basic Usage

Here are some usage guidelines:

Append ". in pixel art style" to your prompt. For example, "cute dragon. in pixel art style".
Downsize the image 8x using nearest neighbor.
Init images are very helpful.
The model performs best at around 128x128 canvas size but is also great for creating smaller items, characters, etc.
Use a VAE with fixed fp16 support: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix.
Do not use a refiner.

Advanced Usage

from diffusers import DiffusionPipeline, UNet2DConditionModel
import torch

pipe = DiffusionPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0",
    torch_dtype=torch.float16,
    unet=UNet2DConditionModel.from_pretrained("pixelparty/pixel-party-xl", torch_dtype=torch.float16),
    use_safetensors=True,
    variant="fp16",
)
pipe.to("cuda")

torch.manual_seed(11215)
prompt = "cute dragon. in pixel art style"
negative_prompt = "mixels. amateur. multiple"

image = pipe(prompt, negative_prompt=negative_prompt, num_inference_steps=25).images[0]

📄 License

Please do not host this model. It is licensed under CreativeML - OpenRail - M.

Property	Details
Model Type	Pixel art - based text - to - image model
Training Data	Not specified
Base Model	stabilityai/stable-diffusion-xl-base-1.0
Tags	text - to - image, stable - diffusion, diffusers
Instance Prompt	. in pixel art style

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご