A collection of tag wildcards that can be used to generate a variety of anime scenes in Stable Diffusion.
Add the Dynamic Prompts extension to your Automatic1111 WebUI (it allows wildcards inside wildcards)
Put the downloaded "wildcards" folder inside the extension's folder
Write __full_scene_plus_character__ in your prompt to start generating random images!
If you have the TagComplete extension you can type double underscore (__) to see a list of all available wildcards. Mix and match them to nail down the exact types of images you want.
You can edit the wildcards however you like from the Wildcard Manager tab!
Description
Updates:
Greatly reduced chance of children
Added more accessories
Added more tags of everything
Improved most tags
Improved accessories and locations
FAQ
Comments (7)
Accidental beauty. Very useful.
ah, someone else who appreciates being able to do full prompts at random. Though I personally use multiple wildcards, rather then one for the whole thing. more modular that way
Thank you and agreed, you can do that with these wildcards too, for example I could use the following for angels:
masterpiece, best quality, anime style, 1girl, solo, __composition__, __locations__, __time__, __poses__ BREAK __colors__ hair, __hair__, __hair_addon__, __colors__ eyes, __body__, {armored dress|white dress}, angel wings, halo, __breasts__, __faces__
is it possible to put LoRAs in the wildcards?
I'm not sure, but probably yes
Yes, I have a file with LORA names and Weight numbers, and then use wildcards like the following to randomize Loras and Weights.
<lora:__Lora__:__Weight__>
easily the best wildcards I've tried so far.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.