LORA

I've trained two sets of clothing. The trigger words are as followstype A: pantyhose, sweater, jackettype B: shirt, skirt, fishnet thighhighsV3.0The previous 1.0 version of Mar...

基于虚拟人物的训练LORA权重：建议为0.8-1强化了触发触发词 xem,1girl,(红眼:1.4),尖耳朵,(恶魔翅膀:1.1),(恶魔角:1.1),(恶魔尾巴:1.1)要出全身加上背带制作分享不易，没有灵感，喜欢的请从...

Sunrise Stance有时候也会产生sunrise的效果（笑）,注意：如果你使用Hires. Fix的话，Denoising需要调低到0.4以下,权重0.5-0.9Sometimes Sunrise Stance can also produce the effect of a ...

Role of Girls' FrontlineWeight 0.8~1.0 ,and you can ADJUST prompt to reductive characterYou can trigger different THREE costumes:1."You Who Takes the Step"：weddingG36C;2.Ode t...

coco bandicoot video game crash work with 0.8 weight enjoy it.Experiment with the scores to get the desired details with the promt:score_9, score_8_up, score_7_up, score_6_up, ...

Trigger: ral-semiconductor☕ Buy me a coffee: https://ko-fi.com/ralfingerai? Join my discord: https://discord.gg/g5Pb8qNUuP 描述: 训练词语: r...

LoRA model for Ryouka art style, recommend strength is 0.5-0.8.Use hires for better result.Add negatives prompt to remove text in background,such as (text:1.4, signature:1.4, w...

Support me onfacebook.com/Kaiseirpatreon.com/Serkaihttps://ko-fi.com/kaiseirWeight: 1.0Trigger words:Appearance:ai hayasaka, blonde hair, blue eyes, blue scrunchie, hair betwee...

接角色训练，根据模型难度决定（88r起步包修改，搜集素材20r）,感谢打赏的朋友，QQ：3250911131，Donate a cup of coffee: https://ko-fi.com/timoplayDonate electricity AFDIAN（爱发电）...

Support me onfacebook.com/Kaiseirpatreon.com/Serkaihttps://ko-fi.com/kaiseirWeight: 1.0Trigger words:Appearance:tearju lunatique, blonde hair, (green eyes:1.5), hair intakes, l...

update 20230419new locon version released,recommend weight 0.6~0.8V1.0这次降低了步数，使得模型没有以前那么硬，可以适当调高权重以获得比较明显的画师风格，推荐权重0.8，模型除了增...

SDXL - Test SFW LoRA CivitAI trainer "modelspec.encoder_layer": "1", "ss_mixed_precision": "bf16", "ss_caption_tag_dropout_rate": "0.0", "ss_dataset_dirs": { ...

TipsIt is trained on Realistic Vision so prefer that, though it should work fine with many other models too.Use highres fix for upscaling because img2img can reduce quality.Che...

I've started a community on Deviant Art to share images made with my models/LoRAs! Join here if you are interested!I'm now on Patreon. Check it out if you want to support my wo...

As requested i tried making a Lora for teeth, i'll upload also the dataset, so if someone is better than me in parametrization can do a better job. Just prompt Teeth or Braces ...

This is one of my LoRa Models and It's definitely not perfect.(as all my LoRA will be)Hope you enjoy anyway!You can try the prompts from the example images!(I'm often using ADe...

Mechanical devices Davinci sketch style 描述: 训练词语: davincisketch ...

This LoRA is intended to give a subtle oil painting feel, so it's better to keep the weight below 1.0 描述: Released 训练词语: oil painting,c...

適用すると光沢感が弱まります,マイナスで適用すると光沢感が強くなります,「deluster1」とは少し出力に違いが出ます,構図への影響を抑えたい場合は、拡張機能の「LoRA Block Weight」を使...

I've trained two sets of clothing. The trigger words are as followstype A: pantyhose, sweater, jackettype B: shirt, skirt, fishnet thighhighsV3.0The previous 1.0 version of Mar...