Empfohlene Prompts

simple but detailed prompt

Empfohlene Negative Prompts

(worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed, underexposed, grayscale, bw, bad photo, bad photography, bad art:1.4)

(watermark, signature, text font, username, error, logo, words, letters, digits, autograph, trademark, name:1.2)

low quality, watermark, ugly, bad anatomy

Empfohlene Parameter

samplers

Euler a, DPM++ 2M Karras, DPM++ SDE Karras

steps

0 - 100

cfg

0 - 20

clip skip

2

resolution

800x1024, 832x1216

other models

WildCardX-XL-Fusion (c539a5c532)

Empfohlene Hires (Hochauflösungs-) Parameter

upscaler

ESRGAN_4x, Lanczos, 4x_foolhardy_Remacri

upscale

1.5 - 2

denoising strength

0.12 - 0.3

Tipps

KEEP BOTH XL FUSION OG & XL FUSION based on your projects.

Less CFG scale is a bit more realistic; higher CFG scale produces crazier results.

Use any sampler but Euler a or DPM++ 2M Karras yield good results.

Use simple but detailed prompts for best output.

Versions-Highlights

Meet the ORIGINAL GANGSTER

WildCardX-XL FUSION OG!!!

Baked XL VAE

Ersteller-Sponsoren

If you feel my work is helpful and useful, Then let's have a Coffee!

TRY IT!!! and you'll see

"WildCardX-XL-Fusion OG"

WildCard-XL-Fusion is finally here! WARNING: NOT YOUR TYPICAL MODEL

If you feel my work is helpful and useful, Then lets have a Coffee!

Whats special?

KEEP BOTH XL FUSION OG & XL FUSION based on your projects.

XL Fusion OG is sharper, with more details, more contrast and richer colors than XL Fusion. Fusion OG and XL Fusion model has highly knowledge of what a ARTISTIC ART photo is. It is a fusion of Artistry and Creativity Mixed with ALL artistic styles. It is a highly trained and properly merged with top models.

How to use?

Prompt: simple but detailed prompt

Negative: "(worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed, underexposed, grayscale, bw, bad photo, bad photography, bad art:1.4), (watermark, signature, text font, username, error, logo, words, letters, digits, autograph, trademark, name:1.2), "
Steps: >20-100 (depending on what style and feeling you want to achieve)

Clip Skip: 1-2
CFG Scale: 3, 5-20 (less is a bit more realistic the higher the CRAZIER)
Sampler: Any Sampler but will give good results with Euler a or DPM++ 2M Karras
Size: 800 x 1024 >
Hires upscaler: ESRGAN_4x,Lanczos, any of your favorite upscaler.

Check out my other Models too

WARNING: UPLOADING/SHARING MY MODELS OUTSIDE CIVITAI IS STRICTLY PROHIBITED. Kindly Let me do it myself.

Checkpoint License : Link

KINDLY Dm me at my Discord: mr.fries1111

Email: im.mr.fries1111@gmail.com

or DM me here at CIVITAI for permissions.

Feel free to leave Reviews and Show your amazing works here - and always have fun creating 🍟🍟🍟

Zurück
NoobAI-XL Detailer - V-Pred1_V1
Weiter
negative_hand Negative Embedding - negative_hand

Modell-Details

Modelltyp

Checkpoint

Basismodell

SDXL 1.0

Modellversion

FUSION OG

Modell-Hash

22ebc61141

Ersteller

Diskussion

Bitte log in um einen Kommentar zu hinterlassen.