modelle/Ri-mix - [PONY + Illustrious] - Ri-mix Ω

Ri-mix - [PONY + Illustrious] - Ri-mix Ω

|
7/19/2025
|
11:24:38 AM
| Discussion

Empfohlene Prompts

1girl, solo, long hair, looking at viewer, masterwork, masterpiece, detailed, best quality, high detail, depth of field, very aesthetic, 8k, dynamic pose, dynamic angle, adult, aged up

Empfohlene Negative Prompts

lowres, worst quality, low quality, bad anatomy, bad hands, jpeg artifacts, signature, watermark, text, logo, artist name, extra digits, censored, patreon username, loli

Empfohlene Parameter

samplers

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

steps

20 - 50

cfg

5 - 7

resolution

768x1152, 768x1152

vae

sdxl_vae.safetensors

other models

0.75(newmix.fp16) + 0.25(dmixfouth) (ff9b5e63a4)

Empfohlene Hires (Hochauflösungs-) Parameter

upscaler

None

upscale

1.5 - 2

steps

10

denoising strength

0.1

Tipps

Process for better results: Text-to-image Highres fix → Image-to-image Face fix → Upscale x2

Use ADetailer with Anzhc Face seg 640 v2 and yolov8n.pt detectors for face refinement.

ADetailer inpaint denoising strength recommended at 0.35 with 20 steps and DPM++ 2M sampler.

Versions-Highlights

Positive and Negative prompt I'm using on ForgeUI works on on-site generation as well

POS - masterwork, masterpiece, best quality, detailed, depth of field, high detail, very aesthetic, 8k, dynamic pose, depth of field, dynamic angle, adult, aged up

NEG - lowres, worst quality, low quality, bad anatomy, bad hands, jpeg artifacts, signature, watermark, text, logo, artist name, extra digits, censored, patreon username, loli

Ersteller-Sponsoren

Lora based on this model - Ri-mix - Style LORA

ADetailer detector - Anzhc Face seg 640 v2 y8n.pt / face_yolov8n.pt

Realistic Illustration style model (UPDATED 6/18/25)

Lora based on this model - Ri-mix - Style LORA

Settings for Civitai Generator

My POS and NEG prompt helps

CFG Scale - 5-7

Sampler - Euler a / DPM++ 2M Karras

Steps - 40-50

Denoise - 0.5

____

Process for a better result

Text-to-image Highres fix >>> Image-to-image Face fix >>> Upscale x2

HMU for if you have better settings

My settings for ForgeUI

Generation

Sampling method - Euler a

Schedule type - Automatic

Sampling steps - 20

Dimensions - 768 x 1152

CFG Scale - 6

Highres. fix

Upscaler - None (works for me)

Hires steps - 10

Denoising strength - 0.1

Upscale by - 1.5

ADetailer

ADetailer detector - Anzhc Face seg 640 v2 y8n.pt / face_yolov8n.pt

Inpaint denoising strength - 0.35

ADetailer steps - 20

ADetailer VAE - sdxl_vae.safetensors

ADetailer sampler - DPM++ 2M

Zurück
Stable Diffusion 3 (SD3) - SD3 Medium
Weiter
Dramatic Lighting Slider - Illustrious - v1.0

Modell-Details

Modelltyp

Checkpoint

Basismodell

Illustrious

Modellversion

Ri-mix Ω

Modell-Hash

f944b4784b

Trainierte Wörter

pos: masterwork, masterpiece, best quality, detailed, depth of field, high detail, very aesthetic
neg: lowres, worst quality, low quality, bad anatomy, bad hands

Ersteller

Diskussion

Bitte log in um einen Kommentar zu hinterlassen.