I made this wildcard pack to test my LoRA <a target="_blank" rel="ugc" href="https://civarchive.com/models/2111003?modelVersionId=2388171">K-Pop Poses - v1.0 </a>and as with the Pose and The Look wildcard packs I thought others might appreciate it. Included in the ZIP file is a directory with a set of wildcards for camera angle, color scheme, environment, pose accessories, etc etc etc. (Same instructions for English and for Danbooru versions)Please leave a thumbs up if you find this (and the other wildcard packs) useful. I do this just for the community here and that's my primary feedback on these since image posts aren't really a thing on Wildcards. The file kpop-pose.txt is the "fire and forget" wildcard callers so you don't have to pick and choose every time. There is a "kpop-pose-prose-workflow.json" or "kpop-pose-danbooru-wildcard-workflow.json" in the root directory. If you use that it with ComfyUI it will save CivitAI ready data into the file. Just load the nodes and enjoy. (Remember that your directory locations and file locations may be different, so check.)In this one the workflow included for the danbooru version includes an upscaler and some info on which upscalers to use for what sorts of things. It works for all of the types of checkpoints but of course, your milage may vary.This is built to work with ComfyUI wildcard support by unzipping the folders into the wildcards directory under app. (If you don't have one yet, just make one and put these in there). If you are using something other than ComfyUI, I wish you luck but have no instructions for you.All of the wildcard settings in the file are PG-13 (ish, the rater is judgmental these days).

K-Pop Poses Wildcards

Use with any model that uses Danbooru prompts. Some of these are on the edge of legit full Danbooru so training milage may vary if you use it with a straight checkpoint.

Files with SHA256: 20318e6ed0145a78570bafbf5123b893a1f2b7d29ecfed8cde7f4d335f61a0cf