Quick Generate
Select a style, write what you want to see and click Generate
What you want in the image.
Clear
Random
* This model is not available for use and might generate different characters compared to the ones seen in the images.
Model Description by Creator
2023.9.20
研究 alpha 在 Loha 画风训练中的影响，并顺便训练了一版模型。
结论：alpha 值越低，模型学习画面中的元素会更趋近概念化（且观察到画风受到了轻微影响，不清楚是不是随机误差导致的），且出图时会尽量保留画面中的原有元素，而不去过多考虑画面和谐度的问题，因此低 alpha 下模型对提示词敏感度高。当 alpha = dim 时，完全参考 Loha 进行出图，并对画面中不和谐的元素进行删减，使画面更加干净，因此高 alpha 下模型对提示词敏感度会较低，而更倾向去考虑如何生成一张好图。
比如这张坐在行李上的洋装女孩，使用 alpha 值较低的模型出图，女孩不仅手上拿着卡牌，背景也镶上了一层卡面：
alpha 的问题大家见仁见智吧，设置的太低，有些次要的提示词 ai 会过度理解，实际使用起来会比较麻烦，设置的高了容易调用，但是对提示词服从度会比较低，具体看训练目标。我这里上传的是 alpha 值较高的版本，使用起来会比较容易。
2023.8.30
使用204张图，优化器采用 prodigy 。
这个自适应优化器完美解决了参数修改太多导致的学习率计算问题，非常适合我这种不会调参的懒狗，强烈推荐。经多次尝试，训练画风时，在较大训练集和 batch size 下，将两个学习率都设置成 1 效果最佳，coef 应设置为2。
其他参数没有过多修改，使用金字塔噪声，另外设置了最小信噪比伽马值为 5 以提高模型泛化性。出图模型均使用 NAI 和 easynegative。
Images Generated With This Model
N/A
Hidden For Public Users
hiten,1girl,solo,aqua eyes,bangs,blue eyes,camisole,cowboy shot,crescent,indoors,window,index finger raised,hair bun,looking at viewer,night sky,ocean,pleated skirt,ribbon,short hair,sitting,smile,water,hair bow,<lora:style_hiten_v2:1>
N/A
Hidden For Public Users
hiten,1girl,solo,blue eyes,holding card,white thighhighs,very long hair,looking at viewer,dress,white skirt,playing card,bangs,full body,blonde hair,white gloves,white headwear,earrings,small breasts,black footwear,floating hair,butterfly,black bowtie,jewelry,hair bow,striped thighhighs,sitting on suitcase,striped bow,juliet sleeves,head tilt,petals,bug,outdoors,swept bangs,half gloves,vertical-striped thighhighs,lo**ta fashion,hat bow,<lora:style_hiten_v2:1>
N/A
Hidden For Public Users
hiten,1girl,solo,long hair,suspenders,holding handbag,outdoors,street,looking back,starry sky,brown hair,very long hair,green eyes,<lora:style_hiten_v2:1>
N/A
Hidden For Public Users
1girl, solo, dress, china dress, chinese clothes, long hair, double bun, hair bun, purple hair, pink eyes, multicolored hair, short sleeves, bangs, white dress, looking at viewer, blurry, blurry background, cowboy shot, medium breasts, ribbon, dutch angle, hand up, floral print, hair ribbon, depth of field, side slit <lora:style_hiten_v1:1>
N/A
Hidden For Public Users
1girl, solo, gift, thighhighs, long hair, green hair, christmas tree, white thighhighs, green eyes, gift box, christmas, looking at viewer, black dress, sitting, fur trim, box, bangs, open mouth, smile, blush, hair between eyes, frills, very long hair, cape, :d, red cape, christmas ornaments, frilled dress, feet out of frame, fur-trimmed cape, holding gift, <lora:style_hiten_v1:1>
N/A
Hidden For Public Users
1girl, little red riding hood (grimm), bird, basket, wolf, hood, twin braids, yellow eyes, nail polish, long hair, tree stump, ribbon, blonde hair, apron, bread, thigh strap, solo, smile, bow, grass, toenail polish, hair ribbon, sandals, scissors <lora:style_hiten_v1:1>
N/A
Hidden For Public Users
1girl, solo, black gloves, holding fork, long hair, green eyes, wide sleeves, standing, cross-laced footwear, earrings, long sleeves, bow, knee boots, high heel boots, green dress, lace-up boots, cake, bangs, black ribbon, striped headwear, vertical stripes, blush, brown hair, apron, green socks, striped bow, green bow, green headwear, detached sleeves, neck ribbon, <lora:style_hiten_v1:1>
N/A
Hidden For Public Users
1girl, blue eyes, solo, armband, flower, smile, red hair, multicolored hair, looking at viewer, open mouth, wide sleeves, bangs, japanese clothes, blush, x hair ornament, :d, holding bouquet, blue butterfly, floral print, frilled apron, kimono, braid, very long hair, day, white apron, leaf, <lora:style_hiten_v1:1>
N/A
Hidden For Public Users
1girl, long hair, solo, white shirt, blue hair, black eyes, blush, apron, chocolate, looking at viewer, ribbon, open mouth, blurry, blurry background, hair ribbon, holding, bangs, mixing bowl, head scarf, valentine, bowl, nose blush, food, long sleeves, sweatdrop, depth of field<lora:style_hiten_v1:1>
Similar Models
Train Models
Train AI Models On Your Own Pictures
You can use your own pictures to train AI models. Reimagine yourself in new ways with our AI-powered image generator.
Create pictures of yourself around the world, in your dream outfit, as a video game character, or in countless art styles. The possibilities are endless.
Train Your Own Model
Generate AI Images. Simple and Easy
AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.