Sign In

Liviel Wildcards for Anime Women

63
869
16
Updated: Mar 7, 2025
toolanimewomanwildcards
Type
Wildcards
Stats
496
0
Reviews
Published
Mar 6, 2025
Base Model
Illustrious
Trigger Words
____RandomFemale/Face/RandomFace____
____RandomFemale/Expressions____
____RandomFemale/Poses/RandomPose____
____RandomFemale/Clothes/RandomClothes____
____RandomFemale/Body/RandomBody____
Hash
AutoV2
0FA22417CB
LI
Liviel

I made these wildcards because every other wildcard sets like these were either too bloated or harder to configure to my liking.

These wildcard are compact, relatively easy to read, and you can easily change the weights to your liking. You can also easily add new options if you want to expand them to have stuff that you like.

The base weights are all based on my personal preferences. I highly recommend going through the files to customize them to your liking. The main purpose of these files is to make them easy to customize.

Some options, like bigger and smallest bust sizes, and certain hair lengths, are at 0 weight by default. I chose this because they are generally more extreme and many people might not like to randomly get them.

You can also just randomize what you need, like eye color or hair length. These wildcards are completely modular in that aspect.

How to install:

You need the dynamic prompts extension for stable diffusion.

Once you have it, place the files inside the wildcards folder (webui\extensions\sd-dynamic-prompts\wildcards)

How to customize:

Each tagset is formated like this:

{

weight1::tags1

|weight2::tags2

|wegith3::tags3

|...}

For example, the Hair/Length is formated this way.

You can only get one tag(s) of the ones inside each set of tags.

Weights are how likely you're to get those tags compared to others. A tag with a weight of 4 is four times more likely to appear than a tag of weight 1.

To add more tags, simply add another set of tags inside the brackets using this format: |weight::tags

Some tagsets are also formated like this:

{weight1::, tags1|weightX::}

{weight2::, tags2|weigthY::}...

For example, the Face/Features is formated like this

They're made this way to make it so you can get any of the tags in that file at a random chance. Sometimes you might get one, sometimes none, and sometimes many or even all.

The likelyhood of the tag making an appearence in the final prompt is calculated as (weight of the first tag/weight of both tags). The higher the 2nd weight compared to the first, the less likely the tag is to come up.

To add more tags, simply add another one following this format: {weight::, tags|weight::}