First time users please carefully read through, it will definitely be helpful!

New version V6 enhanced with traditional Chinese style, for more diversity you can choose V5

Highlights: ssr figure, texture, as well as some composition corrections,

Officially launched, chose a mecha for the cover because it's getting colder and more layers are needed, this model was designed for character design production, it hasn't been released before due to many issues (and of course, there are still many issues now)

Start with 1girl as the descriptor, negative words are based on the cover, also recommend my Lora (used MFlame for the cover image)

https://liblib.ai/modelinfo/55a1423579194b499befaf2c59eefddc?mine=1 , added special effects, results are pretty good

...for full body, use ADetailer, avoid facial modifications

Everyone can refer to the reference image for testing

Compared to previous versions, easier for image output, no need for too many descriptive words, used for assisting in character concept design,provides inspiration for character/environment design, during testingmachine translation content and one-click random word recognition perform quite well,suitable for players who want to create their own characters but lack ideas.Based on the personal version, using a large number of scenes, character postures, and equipment materials for training to improve material and visual performance.

Secondly, model features:

1. From the perspective of game enthusiasts/costume collectors, optimized overall character stiffness issues, can try various character styles

2. Optimized descriptor understanding (model acting dumb)

3. Optimized composition structure, resolves overall character stiffness issues

4. Optimized hand gestures, provides better response to actions and adding visual elements, good compatibility with Lora, refer to example images:

(1) Red-eyed Lora

(2) Cyberpunk: Lucy

(3) Flower in the mirror Lora author:@ Misty River in the Fog

Due to limited personal skills, looking forward to feedback from the pros

Thirdly, based on segmented scenes, image-editing tutorials:

(Example images do not require Lora, control net): (Highly recommend Lao Liu's descriptor plugin: random inspiration is very useful)

(1) Character concept art

Character: Vampire, Dracula style

(2) Character Illustration

And so on for the rest of the content...