Exquisite delicacy | Youke third dimension animation

The core capability lies in the exquisite 3D art style, rich details, realistic and delicate material textures, as well as powerful text comprehension.

Introduction to features:

(1) Exquisite 3D art style: High-quality 3D rendering style is demonstrated in the portrayal of characters, clothing, and various objects.

(2) Rich details: More visual details bring a higher model ceiling.

(3) Realistic and delicate material textures: Can be used for the production of product images, while it can generate related items without lora, but with specific product lora, it can generate more stable outcomes.

(4) Powerful text comprehension: Almost no action overfitting, good support for all terms related to camera angles, perspectives, and concepts.

Cons: When the camera is too far and the scene is too large, the 3D art style is easily lost; cannot create strong men without lora.

Instructions:

1. VAE has been merged: (New users no longer need to worry about grayscale image issues), you can cover the model's own VAE effect by selecting VAE again by yourself;

2. Sampling methods are not very selective, using the common ones should be fine;

3. It is recommended to use high-resolution restoration (Hires.fix) for magnification, with a redraw magnitude of around 0.4, recommended magnification algorithms: 4x-UltraSharp or R-ESRGAN4x+;

4. Different magnification algorithms use Latent, which may cause the entire image to blur, creating a soft light effect. Please determine whether to use it based on actual circumstances. Refer to the difference between two sample images with the same composition (although the sample images may appear to be enhanced, in most cases the effect is not satisfactory).

5. Recommended basic steps are 25-50 steps, high-resolution restoration magnification steps are 10-20 steps;

6. In general, clip1 and clip2 have similar effects, but clip1 is more conducive to maintaining the 3D style.

My other models

Youkengi_realistic(fantasy)SDXL https://www.liblib.art/modelinfo/12fb2246263d46f880829fe4320e89f2

Youkengi_anime_SDXL_base https://www.liblib.art/uploadmodel/4a340e48e0d34d698d0d54d57715b631

Youkengi cute:https://www.liblib.ai/modelinfo/ac8fb054451245778b5ea3452104c731

Youkengi chibi:https://www.liblib.art/modelinfo/33253de83fcb4465a63cd3c421723cf4?mine=1

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

Product image keyword suggestions without lora, XXX can be replaced with your desired content:

Front: XXX_(object), object, (((no humans))), Design, dynamic graphic art, professional design

4k, like photos, product map, white background, blender, C4D, ultrahighres

Negative: ugly, deformed, noisy, blurry, low contrast, animal, human-beings, worst quality, low quality: 1.4, monochrome, low res, fewer digits, blurry, signature

Keyword suggestions are not fixed, after mastering the model usage, you can try adjusting them yourself.

If you are satisfied, please give me more feedback~

Also, by following and bookmarking, you can receive updates on future versions and new models~

No content available