README.md

metadata

license: creativeml-openrail-m
pipeline_tag: text-to-image
library_name: diffusers
tags:
  - art
  - text-to-image
  - stable-diffusion
  - lora
  - diffusers
widget:
  - text: >-
      A refined, noblewoman standing gracefully in a lush garden. She has an
      elaborate updo adorned with delicate pearls and flowers, and she wears a
      flowing, pastel-colored gown with intricate lace details. Her serene
      expression is highlighted by soft, natural lighting. The background
      features blooming roses and a classical stone fountain, creating a sense
      of timeless elegance and tranquility.
    output:
      url: images/image_3.png
  - text: >-
      A bearded inventor with wild, curly hair, standing confidently in his
      steampunk workshop. He wears brass goggles on his forehead, a leather
      apron over a white shirt with rolled-up sleeves, and fingerless gloves.
      His hands are slightly stained with grease, and he holds a small,
      intricate mechanical device. The background is cluttered with gears,
      blueprints, and tools, illuminated by the warm glow of oil lamps, giving
      the scene a creative, industrious atmosphere.
    output:
      url: images/image_4.png
  - text: >-
      A cyberpunk hacker with neon blue hair and cybernetic implants on their
      face, seated in front of multiple holographic screens in a dark, high-tech
      room. They wear a sleek, black leather jacket with glowing circuit
      patterns, and their fingers are covered in cybernetic gloves that
      interface directly with the floating data. The background is a chaotic mix
      of digital code, flashing lights, and wires, emphasizing the high-tech,
      futuristic setting.
    output:
      url: images/image_5.png
base_model: stabilityai/stable-diffusion-xl-base-1.0

cutton_doll_lora-xl

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("./cutton_doll_lora-xl.safetensors", adapter_name="doll_sdxl")

pipe.set_adapters(["lora", "doll_sdxl"], adapter_weights=[1.0, 1.2])
pipe.to(device="cuda", dtype=torch.float16)

prompt = "a chibi doll, cute"
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

Download model

Weights for this model are available in Safetensors format. Download them in the Files & versions tab.

Changelog

v1: Initial release