Pixel Art XL
Consider supporting further research on Patreon or Twitter
Downscale 8 times to get pixel perfect images (use Nearest Neighbors) Use a fixed VAE to avoid artifacts (0.9 or fp16 fix)
Need more performance?
Use it with a LCM Lora!
Use 8 steps and guidance scale of 1.5 1.2 Lora strength for the Pixel Art XL works better
from diffusers import DiffusionPipeline, LCMScheduler
import torch
model_id = "stabilityai/stable-diffusion-xl-base-1.0"
lcm_lora_id = "latent-consistency/lcm-lora-sdxl"
pipe = DiffusionPipeline.from_pretrained(model_id, variant="fp16")
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
pipe.load_lora_weights(lcm_lora_id, adapter_name="lora")
pipe.load_lora_weights("./pixel-art-xl.safetensors", adapter_name="pixel")
pipe.set_adapters(["lora", "pixel"], adapter_weights=[1.0, 1.2])
pipe.to(device="cuda", dtype=torch.float16)
prompt = "pixel, a cute corgi"
negative_prompt = "3d render, realistic"
num_images = 9
for i in range(num_images):
img = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
num_inference_steps=8,
guidance_scale=1.5,
).images[0]
img.save(f"lcm_lora_{i}.png")
Tips:
Don't use refiner
Works great with only 1 text encoder
No style prompt required
No trigger keyword require
Works great with isometric and non-isometric
Works with 0.9 and 1.0
Changelog
v1: Initial release
- Downloads last month
- 22,770
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.