<span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">anima v1.0: <span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">本模型基于 anima_base 训练，专属触发词详见配图。模型对各类 LoRA 适配效果良好。若因 LoRA 或其他因素导致角色还原度不足，可适当调高模型权重，实测最高设置到 1.5。如需还原原生画风，请添加提示词：<code>3d render, jubing style</code><span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">。<hr />以下提示可无视，基于sd1.5，已过时。If you want good generalization, please choose version 1.0.If you want better fidelity and features, please choose version 2.0.Version 3.0 has both characteristics but lacks the strong generalization of 1.0; it can be used on some furry models.V3.0:Utilizing the fluffy rock e27 model as the base membrane for training, suitable for some furry models with good generalization. Similar prompts as below; if Lora is ineffective, try adding prompts such as topless, headband, wolf boy, bandage, etc.V2.0:This Lora is trained for indigoFurryMix V105, with poor generalization. After personal testing, it can generate in furry-related models, but Lora requires different weights, between 0.6 and 1.2, and the effect is not satisfactory. It is recommended to use the indigoFurryMixV105 model for generation with related prompts like xiaoye(character), furry, 1boy, recommended weight 0.8; or use crosskemonoF for generation with prompts like xiaoye(character), furry, 1boy, headband, topless, necklace, bandage, recommended weight 1.1. Adjust the weight to control the character's form using shota prompts, recommended (shota:0.5).V1.0:Recommended weight: 0.6~0.8Recommended trigger words: furry, xiaoye, 1boy, masterpiece, headband, topless, bandage, kemono若想拥有良好的泛化性，请选择1.0版本，若想拥有较好的还原性和特征，请选择2.0版本3.0版本拥有以上两种特点，但没有1.0的良好的泛化性，在部分furry模型上可以使用V3.0：采用了fluffy rock e27模型作为底膜进行训练，可用于部分furry模型，泛化性较好。提示词同下，如果lora无用，请尝试增加提示词，如topless,headband,wolf boy,bandage等V2.0:该lora针对indigoFurryMix V105训练，泛化性较差，经过个人简单测试，可在furry相关模型进行生成，但Lora所需权重不一，0.6-1.2之间，且效果较差。推荐采用indigoFurryMixV105模型进行生成，相关提示词为xiaoye\(character\),furry,1boy，推荐权重0.8；或者crosskemonoF进行生成，相关提示词为xiaoye\(character\),furry,1boy,headband,topless,necklace,bandage，推荐权重1.1，可采用shota提示词调整权重控制人物形态，推荐(shota:0.5)。V1.0:推荐权重：0.6~0.8推荐触发词：furry,xiaoye,1boy,masterpiece,headband,topless,bandage,kemono

小野 (巨兵长城传 ) | Siya Wolf (The Great Warrior Wall)

该模型在crosskemonoF版本上有较好的表现，推荐权重0.7，clip跳过层2

Files with SHA256: 18e8b024f73fb77cd250bfe876c5c08d0c05eed3196336e7ba950aae630caea5