Vanilla (TFF Style)版本V1 (ID: 590656)

Meant to replicate the character "Vanilla" in the art style of TFFeathers. Some issues since the character is not drawn the same way in all the training images (two tone body vs one body colour /leg color /different hair styles) but with the correct negative prompt these problems can be avoided. Generates images well enough to train a more consistent model, so I consider it an OK success. Use at strength 0.9, LoRa might have been overcooked a bit

Check out the other characters and the style LoRa

Example positive:

score_9, score_8_up, against surface, anthro, ass, ass on glass, asshole, avian, beak, hyper ass, hyper butt, breasts, bubble butt, eyebrows, feathers, female, furry, genitals, glass, half-closed eyes, inviting, looking at viewer, looking back, narrowed eyes, naughty face, nude, on glass, open beak, open mouth, pov, pussy, rear view, solo, tail feathers, tongue, voluptuous, colored, digital media \(artwork\),  sexual genital fluids, pussy juice, middle finger, 5 fingers,  <lora:Vanilla_TFF_style:0.9>  

Negative:

Negative prompt: zPDXL2-neg, score_6, score_5, score_4, futa, pubic hair, pubes, body hair, tan lines, bad anatomy, bad proportions, gross proportions, unnatural body, uncoordinated body, disfigured, ugly, deformed, fused bodies, body horror, malformed, mutated, mutation, bad hand, bad hands, bad fingers, mutated fingers, fused bodies, body horror, weird face, ugly face, conjoined, extra arms, bad text, x-ray, signature, young, floating, (ears:1.2), open mouth, short stack, large forehead, nude, holding object, object, bed, animal horns, collar, muscular,  short stack, thin, buff, (wings:1.3),  sagging breasts, messy, spikes \(anatomy\), closed eyes, buckteeth, desert, assymetrical, large hands, child, glasses, url, green body, <3,

Config:

Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 178510165, Size: 1088x960, Model hash: 821aa5537f, Model: autismmixSDXL_autismmixPony, Denoising strength: 0.4, ADetailer model: FacesV1Robust.pt, ADetailer confidence: 0.3, ADetailer dilate erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer model 2nd: hand_yolov8n.pt, ADetailer confidence 2nd: 0.3, ADetailer dilate erode 2nd: 4, ADetailer mask blur 2nd: 4, ADetailer denoising strength 2nd: 0.4, ADetailer inpaint only masked 2nd: True, ADetailer inpaint padding 2nd: 32, ADetailer version: 24.3.5, Lora hashes: "Vanilla_TFF_style: c6a60f4f8632", Mask blur: 4, Inpaint area: Only masked, Masked area padding: 32, TI hashes: "zPDXL2-neg: 29dc1b9b1a2b, zPDXL2-neg: 29dc1b9b1a2b", Version: v1.8.0, Source Identifier: Stable Diffusion web UI

If you like this please upload some images - I get buzz for every image + would love to see what you can do with this!

描述:

Images can be used for LoRa improvement

训练词语: