Gorgeous and lovely | Youkengi cute anime

youkengi cute animeis the model from the Youkengi animation seriesthat represents the "cute" aspect. It is also the most cutting-edge model, simply putit is a hexagonal model，In terms of various parameters, it ranks at the top among existing models，and can achieve great results with arbitrary word stacking.

Its core abilities are:：rich and reasonable character and background details，natural talent，varied and dynamic compositions，no face lock，strong Lora combination ability，very broad image generation range， and the ability to use many tags as Lora，Its ultimate text comprehension（though it hasn't broken through the ceiling of SD1.5, limited by the built-in clip's language comprehension and model parameters). After thorough testing, this model has no obvious shortcomings...

youkengi 3D：https://www.liblib.ai/modelinfo/ad6942b5acbc4301a2a2b30a2b4591ae?mine=1

youkengi delicate：https://www.liblib.ai/modelinfo/cdb5d856b52d49329a481f552d48c4e6?mine=1

youkengi anything：https://www.liblib.ai/modelinfo/b0466fd9edd34ea7a67dab412db47fb3

youkengi clear：https://www.liblib.ai/modelinfo/e5be47b63e5b4408be94942374ebd57c?mine=1

youkengi delicate if： https://www.liblib.ai/modelinfo/19c85fcbdf424520b0fb70613da88ff6?mine=1

This model has incredible potential but is very challenging to master. Be sure to read the instructions below before using it. The "if" version is a special edition of the Delicate model, and a separate page has been created for it.

Usage Instructions:

1.VAE Merged: (No need to worry about gray images for beginners), you can also override the model's built-in VAE effect by selecting another VAE.

2.Sampling Method: Common sampling methods work fine.

3.It's recommended to use high-res fix to enlarge with a redraw strength of around 0.4.，The only not recommended enlargement algorithm is latent, which can blur the image.

4.Recommended base steps is 25~50, 10~20steps for high-res fix enlargement.

5.To balance all parameter capabilities and maintain subtle balance, the model comes with two embedding：negative_hand Negative Embedding,verybadimagenegative_v1.3，which have been tested to not significantly affect image generation range. For beginners, these embeddings are pre-loaded by the lib platform's backend.

6.Two Recommended Experience Tags: Adding "red tone color" at the end allows free tone switching, where "red" can be any color.

（1）Explosive Illustration Background tag：1girl,(((detailed background))), wallpaper,Parting art, mandala,chaos,Radial, streamlined,art design,burst,

（2）Magic Circle Theme Background tag：1girl,detailed background, magic circle, magic circle, magical power,Parting art, mandala,Magic Array

7.Drawing Men: It's recommended to describe the temperament, e.g.,:1man,solo,male focus,Sharp eyes, heroic and handsome,domineering

8.General Negative Words：(worst quality, low quality:1.4), 2girls, bad anatomy, bad hands, cropped, missing fingers,too many fingers, missing arms, long neck, Humpbacked, deformed, disfigured, poorly drawn face, distorted face, mutation, mutated, extra limb, ugly, poorly drawn hands, missing limb, floating limbs, disconnected limbs, malformed hands, out of focus, long body, missing toes, too many toes，monochrome, symbol, text, logo, door frame, window frame, mirror frame

If you're satisfied, please share your feedback!

Also, follow and favorite to receive updates on future versions and new models.~