🎬 Roose_Cinematic_Core_v1

Type: Character LoRA
Base Model: SDXL 1.0
Version: v1
Recommended Weight: 0.85 – 1.0
Trigger Word: roose_cinematic

📖 Description

Roose_Cinematic_Core_v1 is a realistic SDXL character LoRA focused on sharp cinematic portraits, natural skin texture, and a calm, adult presence.

This model was trained as a cinematic portrait core, designed to deliver crisp facial details, believable skin, and consistent identity without overpowering prompts or styles.
It performs especially well in close-ups, studio portraits, and controlled lighting setups.

The emphasis lies on clarity and realism: clean facial structure, expressive eyes, and a grounded, mature look that stays flexible across different prompts.

✨ Key Features

• sharp, high-detail cinematic portraits

• realistic skin texture with natural pores

• balanced, adult facial proportions

• calm and expressive facial presence

• stable results across different prompts

• excellent compatibility with style and lighting LoRAs

• SFW-first orientation

🔑 Trigger Word

Use one trigger only:

roose_cinematic

🧾 Example Prompt (SFW)

(masterpiece, best quality, ultra realistic, raw photo),
roose_cinematic,
1girl,
adult woman,
natural skin texture, visible pores,
calm expression, direct gaze,
cinematic soft lighting,
studio portrait,
85mm lens look, shallow depth of field,
professional photography

🚫 Negative Prompt (recommended)

child, teen, young, underage,
anime, cartoon, illustration,
overprocessed skin, plastic skin,
bad anatomy, extra fingers,
distorted face, low quality, blurry

⚙️ Recommended Settings

• Sampler: DPM++ 2M Karras

• Steps: 28–32

• CFG: 5.5 – 7

• Resolution: 1024×1024 or 832×1216

• VAE: SDXL default

🏷️ Tags

character
woman
realistic
photorealistic
sdxl
portrait
cinematic

📝 Notes

This LoRA is intended as a cinematic portrait character core, not a stylized or exaggerated preset.
For best results, combine it with lighting, background, or clothing LoRAs instead of stacking multiple character LoRAs.