LumenNoir is a cinematic anime-style LoRA trained for Krea 2. It focuses on expressive female portraits, atmospheric lighting, rich color moods, soft cinematic contrast, detailed eyes, and strong visual storytelling.

The style works especially well for warm window light, rainy neon streets, blue night scenes, dramatic backlighting, fantasy environments, and elegant character portraits. It is designed as a flexible style LoRA, not a character LoRA, so the subject, clothing, setting, hair color, and mood can be changed through the prompt.

Recommended starting weight: 0.75–1.0. Increase the weight for a stronger LumenNoir look, or reduce it when you want to preserve more of the base model’s original style.

Best results: use clear natural-language prompts describing the subject, composition, lighting, environment, and mood. The LoRA is intended for cinematic anime illustrations, atmospheric portraits, neon/rain scenes, fantasy characters, and dramatic color grading.

Recommended negative prompt: low quality, blurry, poorly drawn hands, extra fingers, missing fingers, deformed anatomy, duplicate person, distorted face, bad eyes, text, watermark, logo, cropped head.