comic-ink-halftone

SDXL LoRA: Retro Comic Book Ink & Halftone Style

What it does

This LoRA injects a classic comic book style into your artwork, featuring bold black outlines, dramatic cross-hatching, and Ben-Day halftone dot shading for an authentic vintage print look with slight color misregistration.

How to use

Trigger the effect by using 'c0m1cink' in your prompt. A suggested strength of 0.85 typically works well, but you can adjust it based on desired intensity. Use a sampler and settings that work for comic art, such as DDIM at around 30 steps.

Best for

• Superhero scenes
• Fierce dragons and mythical creatures
• Detective stories with mystery elements
• Futuristic cityscapes with robots or mechs
• Pirate adventures on the high seas

Tips

Use punchy primary colors to enhance the comic book feel. Experiment with different strengths of 'c0m1cink' for varying levels of comic stylization.

Example Prompts

c0m1cink A dramatic portrait of a superhero in shadowy alleyway, rain-slicked cobblestones and noir lighting.

c0m1cink An architectural scene with towering skyscrapers in the background, a bustling cityscape at sunset, warm golden light.

c0m1cink Vintage comic book illustration style for an object - a sleek sports car, sharp lines and primary colors against a halftone backdrop.

c0m1cink A dynamic action scene of a hero leaping through the air, dramatic movement with bold ink outlines and vibrant color accents.

c0m1cink Moody comic book page depicting an eerie forest at night, moonlight filtering through branches, shadows and fog creating tension.

c0m1cink Abstract pattern inspired by Ben-Day dots and halftone shading, vivid colors swirling in a psychedelic, retro-futuristic design.

Trigger word

Activate with c0m1cink. Recommended strength: 0.9.

Showcase

Usage with diffusers

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/comic-ink-halftone")
image = pipe("c0m1cink, your prompt", num_inference_steps=30).images[0]