This is a very simple 50/50 merge of the following models:<a target="_blank" rel="ugc" href="https://civarchive.com/models/184813?modelVersionId=329516">AnimeConfettiTuneXL</a><a target="_blank" rel="ugc" href="https://civarchive.com/models/246299?modelVersionId=313717">Comradeship XL</a>It was originally intended for personal use only. I am not an experienced model merger or trainer, so please keep this in mind!It has roots in <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl">Pony Diffusion XL</a>, but was made to work with Danbooru tags thanks to AnimeConfettiTuneXL. I exclusively use Danbooru tags or simple prose in my prompts. I only include the exact terms I want in the positive prompt, and nothing in the negative prompt. However, I do use the negative prompt if I wish to specifically not see something in the generation, or to counter inherent biases of the model (such as "nude" when nudity is not desired). Oh yeah -- this model is tailored toward easy NSFW output! Sometimes censorship is present in the outputs. Adding "censored" to the negative prompt helps with this, but it can change quite a few details if your positive prompt is as simple as I use.For settings, I use:<ul><li>CFG 3</li><li>Steps 20</li><li>Sampler: DPM++ 2M SDE Karras</li></ul>I rarely use this base model on its own. Instead, I use it with my custom style LoRA, <a target="_blank" rel="ugc" href="https://civarchive.com/models/647301/wildswanku-toon-ish-style">WildSwanku</a>. Please see that page for details on how I use them together to create my consistent art style.Feel free to experiment with this base model and show me what you can come up with!

Files with SHA256: 4337d97ecc9598ec1572de4a8ee78c30e662185d4efe4808241abdcbbaef471a