
本合集已停止更新,因为这里上传的都是我初期寻找混合规律的产物,
This collection has ceased to be updated, as everything uploaded here is a product of my initial search for blending rules
描述:
在模型融合时,我最不想发现自己融合的模型与别的模型相似,但我也想让你们知道我融合的结果和使用的素材,来验证我的提取实验。这很矛盾,因此我减少了混合次数以增加素材的辨识度,但这也导致混合的模型缺少差异性
When it comes to model fusion, what I least want to discover is that the model I fused is similar to other models. However, I also want you to know the results of my fusion and the materials I used to validate my extraction experiment. This is contradictory, so I have reduced the number of mixings to increase the distinguishability of the materials, but this also leads to the lack of distinctiveness in the blended model.
Model:https://civitai.com/models/94726
LORA:https://civitai.com/models/119625?modelVersionId=129992
我使用了一个特征模型做减免,你或许能理解这是什么意思
I used a feature model for reduction, perhaps you can understand what this means.
训练词语:
名称: kmMIXByproduct_styleExtractionCute.safetensors
大小 (KB): 2414698
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success