spriton, gelfling, older male, wearing beige brown robes, jewelry, hat,  dark long hair,  half body, mountain range, big eyes,  puppet  <lora:GelflingsV2-000011:1>
stonewood, gelfling, male, ((rian)), wearing leather robes,  armour, dark long hair, fringe,  half body, indoors, puppet , large eyes apart  <lora:GelflingsV2-000011:1>
brea, vapra, gelfling, female, wearing satin lightblue robes,  medium blonde hair, braids, headband, jewelery, half body, indoors, fantasy castle library, smooth, godrays, large eyes apart, wide head, laugh  <lora:GelflingsV2-000011:1>
Quick Generate
Select a style, write what you want to see and click Generate
What you want in the image.
Clear
Random
* This model is available for use and will generate similar character to the ones seen in the images.

Model Description by Creator

EDIT: Well, this V2.0 made it all better :). Happy how it turned out. Settled for Epoch 11.
Please note, prompting has changed slightly, as you can now do characters and clans.

For close or best result use: clanname, gelfling, male or female, type of clothes, type of hair, location, anything else you can think of.
CLAN NAMES: stonewood, sifa, grottan, dousan, drenchen, spriton and vapra.

You can also stick the name of the character in front of their clan name if you wanted Brea, Seladon, Deet, Gurjin or Rian.

RealcartoonRealistic works a bit better with this instead of Photon, clip skip 2
and leave the Lora at 1.0, and you'll be good to go :)


Previous version:


Well.. I donked up! Hahah. Big time. I forgot and pressed the wrong things and it all came out a bit wrong... But, I am sharing the Lora anyway as it is still useful for
creating the odd random Gelfling.

Originally had planned to train all the clans at 768x768 but forgot that too haha, same for the clans they would have tags like 'clanstonewood' or 'clansifa' etc. This did not translate well at all, and the tags won't work now anyway. The only way you can sort of get resemblances of the separate clans is to prompt the actual clan at the start, like 'grottan' or 'vapra', make sure you have the word 'gelfling' and if you want male or female, haircolours and styles (if you can remember them all), that will get you close enough.

O yeah, clothe them also. They be nekkid, so give them some 'robes' at least lol.

All is not lost, at least I can see how it trained according to what I put in the txt files and now I see what I need to revise haha. Use clip skip 1 too as it makes them more puppet like, as clip skip 2 makes them into handsome 3D pixar versions of themselves.

Back to the prompting board I guess!

Images Generated With This Model

Loading...

Similar Models

K/DA All Out Ahri
V1
K/DA All Out Ahri
(0.00)
0
0 Ratings
K/DA All Out Seraphine
V1
K/DA All Out Seraphine
(5.00)
0
1 Ratings
Artoria Pendragon (Swimsuit Ruler) (Fate) LoRA
V1
Artoria Pendragon (Swimsuit Ruler) (Fate) LoRA
(5.00)
0
1 Ratings
Zhongli - LoRA Collection of Trauter's
N/A
V1
Zhongli - LoRA Collection of Trauter's
(5.00)
0
9 Ratings
Princess Zelda LoRA
V1
Princess Zelda LoRA
(4.94)
0
47 Ratings
Diluc -  LoRAs collections of Trauter's
V1
Diluc - LoRAs collections of Trauter's
(5.00)
0
1 Ratings
Kino Character Lora - (Kino no Tabi)
Kino Character Lora - (Kino no Tabi)
(5.00)
0
3 Ratings
Saria
Saria
(0.00)
0
0 Ratings
八雲紫 Yakumo Yukari
八雲紫 Yakumo Yukari
(4.00)
0
1 Ratings
Moomin - LORA
Moomin - LORA
(4.00)
0
1 Ratings

Train Models

Train AI Models On Your Own Pictures

You can use your own pictures to train AI models. Reimagine yourself in new ways with our AI-powered image generator.
Create pictures of yourself around the world, in your dream outfit, as a video game character, or in countless art styles. The possibilities are endless.
Train Your Own Model

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.