This LoRA captures the raw, kinetic, and pencil-heavy art style of comic book artist Cary Nord. It is trained to replicate his signature "direct-to-color" aesthetic, where the roughness of the pencil lines remains visible, giving generations a gritty, organic, and highly expressive feel. And yes, I am using AI to generate a description because I'm lazy and I know nobody reads this shit. Cry about it, I don't care. Either way, I hope this comes out well and people like it. If not, oh well.What This Model DoesThis model applies a distinct "sketchy" texture to your generations. It moves away from smooth, plastic-looking digital art and leans into:<ul><li>Heavy Pencil Textures: Visible cross-hatching and graphite textures.</li><li>Dynamic Anatomy: Exaggerated musculature and movement typical of modern dynamic comics.</li><li>Atmospheric Lighting: High-contrast shadowing often found in noir or dark fantasy settings.</li></ul>What It Is Good At (Best Use Cases)<ul><li>Dark Fantasy &amp; Barbarians: Naturally, this model shines with Conan-esque themes—fur, steel, grime, and battlefields.</li><li>Gritty Superheroes: Think street-level heroes (like Daredevil) rather than bright, glossy capes.</li><li>Action Scenes: Nord’s style is fluid; this LoRA handles mid-combat poses well without looking stiff.</li><li>Sketch/Traditional Media styles: excellent for prompting "pencil sketch" or "graphite" artworks.</li></ul>What It Is Bad At (Limitations)<ul><li>Clean/Vector Art: If you want flat colors, thick clean lines, or a "corporate vector" look, this model will fight you.</li><li>Anime/Kawaii: It struggles with the large eyes and smooth skin typical of anime models; it will try to add grit and muscle definition where you might not want it.</li><li>Photorealism: While it handles lighting well, the texture is inherently illustrative.</li></ul><hr /><h3 id="recommended-settings">Recommended Settings</h3><ul><li>Trigger Word: <code>crynrd-eng</code></li><li>LoRA Weight: 0.6 - 0.8 (At 1.0, the pencil texture may become too noisy; lower it to mix with other styles).</li><li>CFG Scale: 4 - 7</li><li>Clip Skip: 2 (Standard for most anime/comic checkpoints)</li></ul><h3 id="example-tags-to-try">Example Tags to Try</h3>(Use these to help activate the style)rough_sketch, pencil_(medium), traditional_media, Also, try removing quality tags if things are coming out too clean. Also, I found best results with a CFG of 7 and tweaking the LoRA strength. It can take some finessing but I think there's some great potential with this LoRA.<hr /><h3 id="how-not-to-use-it">How Not To Use It</h3>Don't use this LoRA if you are trying to generate "clean" line art for coloring books. The model introduces a lot of inherent "noise" (pencil shading) that makes the image look finished and textured, which is difficult to color flatly.

Files with SHA256: a2a667764e1a2f1c7103ae0dcf04ab8da14163afee4deeb54784aeeeec9517c2