AnythingQingMix - v30
精選圖片
推薦反向提示詞
bad anatomy,text,low quality
推薦參數
samplers
steps
cfg
clip skip
resolution
vae
other models
推薦高解析度參數
upscaler
upscale
denoising strength
提示
使用V3以獲得更優秀的標籤準確度、光影效果及真實感。
謹慎調整標籤準確度;若V3較難上手,可嘗試V2或V1。
加入品質相關標籤以生成更逼真且細緻的影像。
生成圖像時避免使用面部修復。
嘗試使用高分辨率修復以獲得更佳效果。
創作者贊助
(我的QQ群:235392155,Lora代練、ckpt融合調整加我qq:2402799912)
我在國內站tusi.art和liblibai.com裡也都有模型,按平台的創作激勵會有所不同,還請多多關注支持!(我會在此模型的示例圖中公示一部分的圖出來)
【閒魚】https://m.tb.cn/h.5V7ITvv?tk=eY01dB4UcnH
這是我的愛發電,補貼一下電費QAQ,感謝您的支持~
禁止使用此模型進行一切商務及違法行為,禁止隨意轉載,僅作成果分享,違者後果自負!
(我的QQ群:235392155,Lora代練、ckpt融合調整加我qq:2402799912)
我在國內站tusi.art和liblibai.com裡也都有模型,按平台的創作激勵會有所不同,還請多多關注支持!(我會在此模型的示例圖中公示一部分的圖出來)
【閒魚】https://m.tb.cn/h.5V7ITvv?tk=eY01dB4UcnH
這是我的愛發電,補貼一下電費QAQ,感謝您的支持~
一、模型特徵概述
V3:
一、更新操作:
1、利用MBW插件,通過窮舉法融合了部分basil mix的輸入層和中間層
2、利用MBW插件,通過窮舉法以非常低的權重融合了部分模型的輸出層
二、更新效果:
1、提高了tag的準確度
2、提高了飽和度,增強了畫面的真實質感
3、提高了光影效果
4、提高了肢體的準確率
5、提高了角色lora的還原度
三、注意事項
1、建議使用vae:animevae,防止過度飽和(畫廊示例圖分別為此vae和84000)
2、對tag把握更準確也意味著對tag的書寫要求會更高,若V3難上手可以嘗試V2或V1版本
V2:
(若只喜歡成熟的肌肉男性或希望使用更簡單可以嘗試V1版本)
一、更新操作:
1、將NovelAI原版VAE替換了模型內VAE
2、窮舉法更換了ckpt內的clip模型
3、與V1未融合lora版本進行融合
二、更新效果:
1、提升了tag的準確度
2、提升了構圖質量
3、減輕了融合lora導致的形象固定
4、減輕了對於部分tag的過高權重
V1:
1、泛化型融合模型,能對應tag跑出多種畫風,具有很高的跑圖下限、tag準確性,並且從返圖區還能看到他一大能力^v^
2、對人體塑造能力很強,肢體至今未遇到崩壞情況,手部、腳部崩壞機率很小
3、融合時特意避開對臉部的影響,所以臉部沒有固化,很適合搭配角色lora
4、無clip偏移問題
5、因為unet有些過擬合導致某些tag不聽使喚
二、跑圖建議
(具體可以看封面圖片的各類參數,不同VAE有不同的效果,我還喜歡開Face Editor插件)
1、因為融了相當一部分真人模,使用跑圖時可以嘗試和跑真人底膜一樣的參數,但依然非常不建議使用面部修復
2、可以在跑圖時加點質量詞,它們是有用的。如果你希望他看起來更逼真立體,用一些與寫實、光影相關的tag
3、可以多多嘗試在同一tag下,clip跳過層為1和2的實際出圖情況
4、方圖、長圖、寬圖表現都不差,可以隨意調整畫布大小
5、嘗試使用高分辨率修復,不建議使用面部修復
6、請多多點讚、返圖、評論、5星哦~
三、融合採用模型
(我無法確定是否是這些,僅當參考)
LORA:(總共權重為0.3)
Animale LoRa - Animale LoRa v1 | Stable Diffusion LoRA | Civitai
AIroticArt's Penis Model (LoRA) - v1.0 LoRA | Stable Diffusion LoRA | Civitai
CKPT:
AbyssOrangeMix2 - NSFW - AbyssOrangeMix2_nsfw | Stable Diffusion Checkpoint | Civitai
maturemalemix - v1.2 | Stable Diffusion Checkpoint | Civitai
YaoiGen - YaoiGen v0.4.4 | Stable Diffusion Checkpoint | Civitai
Plazm Men - Plazm v1.0 | Stable Diffusion Checkpoint | Civitai
HomoDiffusion (gay) - Homo Diffusion v1.0 FP32 | Stable Diffusion Checkpoint | Civitai
四、給予訪客使用模型的一些建議與回答
(English translation below)
(作為一個外行人,雖然沒有經歷過美術和ai的專業課學習,但我仍然在不斷努力學習探索,並與各位大佬交流心得。我對自己在意的事情有著很高的精神潔癖,因此我希望能在自己的一畝三分地裡傳達正確的ai知識。這些在下方文檔裡都有,僅作為遇到的問題單獨回答。)
1、LORA
對於動漫角色lora而言,最好的權重便是1。雖然我們能通過改變權重的方式來改善擬合,但改變權重會額外產生一些不好影響。比如:c站動漫角色lora大多數都過擬合,而為了改善這一現象往往會推薦降低權重到0.6/0.8,而這一舉動會導致角色失去部分原本的外貌特徵
過擬合通俗解釋:過分訓練導致的lora表現僵化,如人物不聽tag話甚至出原本素材的圖
2、CKP
clip偏移和unet過擬合都會導致模型不聽tag話
clip偏移會導致tag識別問題,c站有相當一部分模型存在此問題而不自知,如果你感興趣,在下面的“其他”裡我會給予檢查及簡單修復的途徑
unet過擬合也會讓模型僵化,不聽tag使喚,甚至在無tag的情況下跑出好看的圖片,這也是過擬合吐原圖的情況
融合類ckpt能夠很好的提高模型出圖的下限,但融合模型對於tag的權重是很亂的,因此會各具“特色”,同樣,基於此類模型訓練的lora很難在其他模型上適用
如果你希望大模型對於lora及其他模型的接納程度更好就盡量不要再融合ckp模型時融合lora,或者和我一樣在權衡之下融入以較低權重修正不盡人意的部分
3、VAE
CKP自身就有一個VAE,外挂VAE不是用於額外的增加,而是用於替換
VAE只是最直觀的是飽和度變化,這並非VAE功能的全部,就對於跑圖而言還會改變構圖、細節等等
4、其他
模型佔內存大的不一定好,很多模型內有一堆無用數據,下載完全浪費流量內存
示例圖的好壞不能確定模型質量,除了基於作者本身審美外,你無法確定他用了多少模型、插件的輔助,你也無法確定他文生圖跑了幾次、圖生圖跑了幾次
下載量與點讚也不能確定模型質量,點讚量基於下載量,而下載量會和作者名氣、封面圖片受眾、模型角色人氣、模型畫風受眾、模型類型受眾等等相關
模型並非版本最新為最好,很多時候是基於某一版本的不同方向的調整。甚至有部分作者為了通過更新的推送而蹭到更多的下載量而惡意刷更新,實際使用下來模型的質量不過是在原地迂迴踏步
其他更為專業詳細的知識請看萬象熔爐 | Anything V5/Ink的簡介部分,關於clip偏移的檢驗修復的擴展連結在此文檔前段部分中
模型連結:萬象熔爐 | Anything V5/Ink - V3.2++[ink] | Stable Diffusion Checkpoint | Civitai
4. Suggestions and Answers for Model Usage by Visitors:
(As a layperson without formal education in art and AI, though I have been continuously learning and exploring, and exchanging experiences with various experts. I hold a strong commitment to accuracy in matters that concern me, and thus I aim to convey correct AI knowledge in my space. The details are found in the document below and are provided only for separately addressing encountered issues.)
1. LORA:
For anime character LORA, the ideal weight is 1. While we can improve fitting by adjusting weights, this can have additional undesirable effects. For instance: On certain image-sharing sites, many anime character LORAs are overfitted. To mitigate this, weight reduction to 0.6/0.8 is often recommended. However, this may result in characters losing some of their original features.
Simplified explanation of overfitting: Excessive training leads to stiff LORA performance, causing characters to not adhere to tags, or even generating images unrelated to the input material.
2. CKP:
Both clip offset and unet overfitting can cause models to not adhere to tags.
Clip offset leads to tag recognition issues. Many models on certain platforms have this problem unknowingly. If interested, I will provide ways to check and repair this in the "Other" section below.
Unet overfitting can also result in stiffness, not following tags, or even generating appealing images without any tags. This situation is when the overfitting produces images that are nearly identical to the original material.
Fusion-style ckpt models can enhance the lower limit of image generation quality. However, the tag weights in fusion models tend to be disorganized, resulting in diverse "characteristics." Similarly, LORAs trained based on such models are less compatible with other models.
If you want larger models to integrate LORA and other models better, avoid fusing LORA when merging ckp models. Alternatively, consider, as I have, incorporating underwhelming parts with lower weights in your fusion process.
3. VAE:
CKP itself has a VAE. External VAEs are meant to replace, not supplement, it.
The most immediate effect of a VAE is saturation alteration. However, this is not the entirety of its function; it can also impact composition, details, and more when generating images.
4. Other:
Models with large memory consumption aren't necessarily superior; many contain extraneous data, wasting bandwidth and memory.
The quality of sample images doesn't definitively determine model quality. Besides the author's aesthetic preferences, you cannot ascertain how many models, plugins, and iterations were involved. Download counts and likes are not definitive measures either; likes are often linked to downloads, which are influenced by the author's reputation, cover image appeal, character popularity, art style compatibility, and audience type.
Newer versions of models are not always superior. Often, adjustments are made based on different directions within a specific version. In fact, some authors maliciously exploit updates for increased downloads. In practice, the quality of the model only treads in place.
For more specialized and detailed knowledge, please refer to the introduction section of "萬象熔爐 | Anything V5/Ink." Expanded links for examining and repairing clip offset issues are located in the earlier part of this document.