Details
Empfehlungen
v2.0
v1.0
Macromastia - Tiffany for Flux

Macromastia - Tiffany for Flux

58
7
83
#Charakter
#asiatisch
#Gorgeous
#
#Gnaden
#asian woman
#Macromastia
#Gigantomastie
#FLUX

Asian Macromastia series

1/12/25 update: v2 was trained on 42 images of her real body with high-resolution images of her 50cm long macromastia and body, including her . This version generates much better of her and . Again, both her and require prompting in order to get accurate generations

This is a macromastia lora. Basic prompting will generate her body with normal since Flux does not want to show you the true Lora. You need to specifically prompt for things like giant , long in order to trigger those results. She also has very large , which Flux does not understand, so you are better off trying to use a Lora or prompting for small .

I cannot post samples since it will violate ToS

Tiffany is a gorgeous model that was trained from clothed and including 48 headshots, 6 body photos, and 9 macromastia . Play around and see what you get. I think I used a hair lora and makeup lora on some of the posted photos, but she was trained with black hair. Changing hair color seems to be pretty easy.

This Lora was created with oversized macromastia (36R) down to her and works pretty well with and without clothing. My best results were clothed. Flux seems to randomly vary the length in a good way, providing a good range. Better prompting helps zero in on what you are looking to achieve.

For some reason Cumzilla v2 works INCREDIBLE with this Lora and not my other Loras. The results are truly amazing. Give that a try!

Training details:

  • Steps: 4,000, 2 batch (ai-toolkit)

  • Trigger word: tiffany

  • Linear/Alpha 32

  • Learning rate: 0.0001 (bf16)

  • Image size: 1024x102>

  • Dataset: 54 detailed photos of her face, head, body proportions, and body

  • Training prompts: Selfies and photos with friends. Makeup in 30% of the dataset. The lora was also trained on very specific beauty marks and skin imperfections, those came across very well.

Comfy settings:

  • Steps: 35

  • Guidance: 3.5

  • Sampler: DPM++ 2M, Beta

I generated 60 images with this Lora and was happy with the results and posted this. Please reach out with any questions or want me to send over any specific images.

Get to know Tiffany:

  • Age: 34, born and raised in America

  • Languages: 普通话, English, French

  • Measurements: 152-79-109 (60-31-43 US), shoe size EU35 (US5)

  • Height: 147cm (4'10), Weight: 71kg (156lbs)

  • Hair: Black

  • size is 34X, size is XL

Übersetzung anzeigen

Bewertungen & Kommentare

-- /5
0 Bewertungen

Noch nicht genügend Bewertungen oder Rezensionen erhalten

no-data
Keine Daten
W
wodz30
11
248
Chatte mit dem Modell
Mitteilung
2024-10-28
Modell veröffentlichen
2025-01-12
Modellinformationen aktualisieren
Modelldetails
Typ
LORA
Veröffentlichungsdatum
2025-01-12
Basismodelle
Flux.1 D
Auslösewörter
t1ff4ny
Kopieren
Umfang der Lizenz
Source-Modell: civitai

1. Dieses Modell dient nur Lernzwecken. Urheber- und Auslegungsrechte liegen beim Originalautor.

2. Bist du der Originalautor eines Modells, kontaktiere uns bitte zur Authentifizierung über unsere offiziellen Kanäle. Wir schützen die Rechte aller Schöpfer. Hier klicken, um es zu verifizieren.

Umfang der Kreativlizenz
Online Generierung
Fusion durchführen
Download zulassen
Umfang der Business-Lizenz
Generierte Bilder können verkauft oder für kommerzielle Zwecke verwendet werden
Modelle können weiterverkauft oder nach der Fusion verkauft werden
QR Code
SeaArt App herunterladen
Setze deine KI-Kreativreise auf dem Mobilgerät fort