//=time() ?>
From 90 to 126k it still produces poor results, but it may be workable. I'll try retraining anyway, I guess. REQUIRES a CFG scale lower than 10 and calling the embedding near the end of the prompt to be adequate.
https://t.co/3n45LipS0E
#幽香 #AI #textualinversion https://t.co/b8Jh5SPH5U
Gundam X's dialogue about Newtypes fascinates me, since it approaches it from a metatextual sense to touch upon how the allegory has been interpreted by viewers.
X is a story of disambiguation & clarity, upholding the qualities Newtypes represent while moving ahead without them.
added the lashtalv2 embedding to my ko-fi as a download for members! finetuned via Textual Inversion with #StableDiffusion2, love how this one turned out ✨
Link below
↓↓↓
Check out my re-contextualization of one of Raja Ravi Varma's classics on @foundation
https://t.co/9BQduyj31S
@Mo_htth @KaiserCrimson_1 @Ludedwolf @RobKuro85 @RichardElric2 "All KAME AMERICA is nothing more than a patriotic color ripoff of MS MARVEL + power rangers"
+sailor moon
+the Classical elements
+the three enduring graces
Kamen America is intertextual AF.
#StableDiffusion2 & textual inversion. 2.0 really does seem great for customization
Cyclizar is the funniest fucking concept imaginable. a Pokémon that is entirely unspecial, doesn't evolve to or from anything, but also, it's textually related to *both* of the box legendaries
you have Koraidon, the "Winged King;" Miraidon, the "Iron Serpent;" and also Dennis
@Canelo No canelo tu no eres así!!!
No le hagas caso a un vídeo descontextualizado 😭😭👊
Despite a very consistent style, the results still fail to match it most of the time at 78k. Besides hair, it also cannot discern the requested color of the vest, which is rather alarming. I may have to retrain it from scratch.
#幽香 #AI #textualinversion https://t.co/FJbHkWKYbZ
66k steps with last 10k done at 1.0e-04 learning speed, getting closer and more consistent now.
#幽香 #AI #textualinversion https://t.co/UEs1bQMVBJ
asutora_gs31800 -> no embedding -> miata8674_gs45000 -> atawashi_gs56000
#幽香 #AI #textualinversion #portrait https://t.co/GDA22hOKdt
By 56k steps, it still hasn't learned much but the coloring method, and the results are mostly a mess.
#幽香 #AI #textualinversion https://t.co/0g3ZlCfJKW
Tão romântico, nem parece que a descrição textual dessa cena faz minha pressão baixar
Training continues, 33k steps now. For some reason it keeps trying to give her yellow hair. The artifacts are severe. Still quite a ways till it's passable even at CFG 8. Fourth pic is with the style invoked last, the rest are early in the prompt.
#幽香 #AI #textualinversion https://t.co/zDB4ifodHV
"I thought I'd try WD with just a textual inversion of myself, this result is too funny and also kind of nice. To all the waifus, hmu!"
#StableDiffusion #AIArt #AIArtwork #DreamStudio
#anythingv3
1 枚の画像からだけでも十分に高精度な Embeddings を作成できるという次世代 Textual Inversion 実装の Dream Artist を試してみた
https://t.co/QXAJA0Oz5Y
素材はランスシリーズ、シィルさんの公式ページの立ち絵
学習回数 4000 Epoch 程度でこの精度よ
生成画像 ←→ 学習元画像
Kiriko from Overwatch 2
#waifudiffusion 1.3 #textualinversion embedding
>100 source images, 50k steps
just paste the file into the embeddings folder of the webui and prompt with "kiriko"
https://t.co/FUSoSs518K
Same parameters but different styles invoked in the prompt, clockwise from top left:
No embedding —> miata8674_gs45000 ⮒
asutora_gs31800 —> (training-in-progress) atawashi_gs22000.
#幽香 #AI #textualinversion https://t.co/ohCrjIRRbL