モデル/Pony Realism Slider - v1.0

Pony Realism Slider - v1.0

12/15/2025

5:05:48 PM

推奨プロンプト

sharp eyes,perfect teeth,realistic hands

score_9, score_8_up, score_7_up, BREAK 1girl, close-up, soft blush on cheeks, symmetrical features, long eyelashes, arched eyebrows, subtle and natural makeup, delicate jawline

推奨ネガティブプロンプト

avoid score_6,avoid score_5,avoid score_4

Text, watermark, blurry face, bad quality, bad hands

推奨パラメータ

samplers

DPM++ SDE, DPM++ 2M

steps

cfg

clip skip

resolution

832x1216

vae

sdxl_vae.safetensors

other models

Pony Realism Slider (b6b12849a476), cyberrealisticPony_v65 (f42e486d3f), Dramatic_Lighting_v3_alpha1.5_rank6_noxattn_100steps (c7fc2cd4c9ee), detail_slider_v2_alpha1.25_rank4_noxattn_100steps (f0414ae9413e)

推奨ハイレゾパラメータ

upscaler

R-ESRGAN 4x+

upscale

steps

denoising strength

0.1

ヒント

Works seamlessly in 0.10 increments for fine-tuned control over detail enhancements.

Best results between +1.0 and +3.0 for clean, natural upgrades to image detail.

Avoid stacking too many quality-related tags in your prompts for balanced results.

Avoid using tags like score_6, score_5, or score_4 in negative prompts to maintain consistency.

Limit the use of other LORAs simultaneously to prevent overcooked results.

バージョンのハイライト

Avoid using score tags (e.g., score_6, score_5) in negatives. No embeddings needed—slider works best standalone. Focus your prompt for optimal results.

クリエイタースポンサー

Come hang out, test LoRAs 🧪, give feedback 🗣️, see WIPs 🎨, and more. 👉 Join here: https://discord.gg/WeqYRZudhW

Pony Realism Detail Slider V1

This slider is crafted to enhance image quality and realism across SDXL models like PonyXL, Illustrious, and other compatible bases. The Pony Realism Detail Slider focuses on improving overall image clarity, texture, and realism, while leaving smaller details like eyes and teeth untouched unless explicitly prompted. It's perfect for achieving natural, polished results with minimal effort.

Key Features:

Works seamlessly in 0.10 increments for fine-tuned control over detail enhancements.
Best results between +1.0 and +3.0 for clean, natural upgrades to image detail. Going beyond +3.0 can add stylistic changes but may slightly alter the subject's appearance.
While this slider enhances overall realism, tiny features like eyes, teeth, or intricate accessories require separate prompts for the best results.
No activation prompt required—simply adjust the slider to control realism and detail effortlessly.
Avoid stacking too many quality-related tags in your prompts. A minimal approach delivers better and more balanced results.

If you need sharper eyes or teeth, add specific prompts such as "sharp eyes," "perfect teeth," or "realistic hands." During the testing, the LORA overdid the eye and teeth details from time to time. This is easily mitigated with negative prompts.

Avoid using tags like score_6, score_5, or score_4 in negative prompts, as they may reduce overall consistency.
To maintain a natural look, limit the use of other LORAs simultaneously to prevent overcooked results.

Enjoy the slider?
Leave a like, share your creations, and drop your feedback! I'm always open to new ideas and requests—your input fuels the creativity and helps improve tools like this.