VRChat Furries

VRChat FurriesDiscord server to see or share other generated images: Discord LinkNew version : V4.2Trained for the Yiffymix v3.3 model (https://civitai.com/models/3671?modelVersionId=143769)YOU NEED TO REDUCE THE STRENGTH OF THE MODEL TO <0.8Work better with CFG Scale between 3 and 6Works with clip skip 1 and 2Contains :Rexouium (keyword: rexouium)Avali with kitavali and davali (keywords: avali, kitavali or avali, davali)Nardoragon (keyword: nardo)Deuzear (keyword: deuzear)Wickerbeast (keyword: wickerbeast)Protogen (keyword: protogen)Taidum (keyword: taidum)Awtter (keyword: awtter)Also works well with anime boy and anime girlLoRAs working well with VRCFurryGacha splash LORA v3.1 and v4 : painting style + epic background and arrangements + improves faces (biased towards shorts + thinner body + centered subject) https://civitai.com/models/13090?modelVersionId=26536HD Helper : avoid duplicate faces on high resolution pictures (https://civitai.com/models/110071?modelVersionId=118644)Fluffyrock Quality Tags : slightly improve characters + more likely to generate furry instead of humans (https://civitai.com/models/127533?modelVersionId=151790)Micro Waist : self descriptive (https://civitai.com/models/4602/waist-slider-microwaist)Negative Embedding working well with VRCFurryI tried a bunch of different negative embedding and those are the ones that seems to give the best resultsBoring_e621 (https://civitai.com/models/87781/boringe621-negative-embedding-improve-image-quality)veryBadImageNegative (https://civitai.com/models/11772/verybadimagenegative)BadDream (https://civitai.com/models/72437?modelVersionId=77169)EasyNegative (https://civitai.com/models/7808?modelVersionId=9208)I recommend to use all of them à the same time.KeywordsImprovers :good lights : makes volumes pops moregood colors : better color choicesgood composition : more assertive composition choicesgood face : faces I preferbestart : used this keyword on the best art of my dataset (few example so it is pretty weak/random)aestheticcartoon : more assertive colors and texture choicesaestheticmuscles : targets no fat, no steroid or bara look (sometimes the position gets a bit stiff)musculegs : to counter act the strong biasis of most models toward thick upper body and thin legsWorsener : (to use in the negative prompt)bad face, ugly, dumb face : used in dataset, pretty self descriptivebad hands : used in dataset, pretty self descriptivebad muscles : unaesthetic muscle insertions, poor definition or disproportionnedbad abs : same but specific to absghbelly : used in the dataset for muscular bodies with gh/steroid bubble gut, and distended absthin legs : self descriptivesmall eyes : self descriptive (but pretty weak as there are very few examples in the dataset)Non vrcfurry keywords but that I like to use with this modelgorgeous eyes : improve eyesextremely detailed furs, 4k, 8k : add details but often biased towards small eyesStyles (keywords) :comicthick linesphotographyrenderpaintingswaggydrafthardstyledetailed styleComposition (keywords):rim lightdynamic scenecomplex scenebicolormonocolormorningBody / Face (keywords):cuteteenager( - / slightly/ very / extremely) muscularripped muscles(spiky / wavy / feather) hairsrelaxedvoreClothes (keywords):( - / tight) shirtshortlegginghoodiebandanacroppedhoodiemuscles through clothesPosition (keywords):side view3/4 view, three-quarter viewback viewlooking backlooking downlooking awaydefying lookdynamic positionseated / sittingcrouchinglying onleaning onfacing downSuggestive / NSFW (keywords):dick through clothesdick sticking outautosuckingjacko posegrabbing pecfirst person viewOthers (keywords):hmoma : furry with human (at least 2 persons)Tips & tricks:Improve avali look by adding those keywords: cute, big eyes, feathers, black eyesGet better faces by adding these keywords: close up portraitSometimes you can get better faces by changing hairs / furs colorsGood positive prompt keywords :Kind of a meta I found across all my modelsgood lights, good colors, good composition, monocolor, bicolor, good face, (cute:0.7), teenager, good lights, good colors, good composition, are keywords I added on the best artworks in the datasetmonocolor, bicolor, are often associated to higher quality / artistic artworks in the dataset with interesting color management(cute:0.7), teenager, cute put emphasis on eyes and make them much better and teenager couteract the "chibi"/"kid" look it could create while still keeping a cute lookand some bonus keywords that can make it look even better, even if it depends a lot on your tastes ^^masterpiece, extremely detailed furs, 8k, soft fur, perfect composition, detailed shading, dynamic pattern with accents, detailed style, soft,Good negative prompt keywords :(neck fluff, fluffy neck,:1.2), lores, bad anatomy, warped, bad hands, BadDream, easynegative, boring_e621_v4, (ghbelly, bad face, ugly, dumb face, thin legs, flat hairs, bad abs, fluffy neck, neck fluff, fluff on the neck:1.2), (anime boy, human, human ear:1.4), verybadimagenegative_v1.3There seems to be a bias towards females, so if you want to generate males, use:female, woman, girl, 1girl,If the shirts are cropped / always lifted, use:crop top, cropped shirt, bra, shirt lift, Planned for next version (V5):Better tail controlReference sheet addition ?better dataset captionningI don't really know, don't hesitate to send ideasno mamagen please