這是我的第一個訓練作品，由於經驗不足，導致訓練材料已經跟實際出圖狀況有不小差異。LoRA明顯有學習到精緻髮型及髮飾、有特色衣服跟嬌小的身材，缺點是過於2D感的五官，過於白皙的臉蛋，以及容易畫錯的眼睫毛等，全身照更容易顯露出缺點，但採用高清出圖似乎狀況較佳。寫實模型可以利用ADetailer來改善這些臉上的毛病，動漫模型則問題幾乎看不出來。取名古典混血，是希望它能提高混血跟五官的質感，但又不會太過於西化，保留一些婉約的東方氣質。
寫實模型建議權重大概是0.3~0.5，2D動漫模型建議權重大概是0.6~1，目前都是參考值。
我最喜歡寫實模型是Asian Realism Fusion、SicaProMax、SimpleMix V3，其他只要不是太寫實或歐美比例的模型大概都可以嘗試它。
常見模型參考: Chilloutmix、麥橘6、XXMix_9realistic
其他模型參考: Real_AliceMix, Meris Realistic（TA）, Only Realistic, Dreamy Remix（TA）, Vigee-Asian_Realistic （TA）, CampurSari（TA）, AWPortrait,
我習慣用的參數:
Size: 682x1024
Steps: 30, Sampler: DPM++ 2M Karras, CFG scale: 7.0,
Vae-ft-mse-840000-ema-pruned
11/23
ADetailer建議重繪幅度0.4~0.5，才能達到較佳唯美效果，並盡量調整與LoRA外觀相似的年齡。
Adetailer Prompt: (((ulzzang belle))), (((pitiful eyes))), ((pretty lips)), ((gentle smile)), (shy face), Adetailer Negative: (((lowres, 2.5D render, CG texture, unreal engine))), (((low quality, worst quality))), (((makeup, eyebrows, eyelashes, lipstick:1.4))),
Description
FAQ
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.