此模型的原作者为「@IronCatMan」，LiblibAI 致力于推动原创模型的分享和交流，因此转载了该模型、供大家非商用地交流、学习使用。请使用该模型的用户，遵守原作者所声明的使用许可。

如果您是该模型的原作者，请与我们联系（邮箱：liblibai@163.com；微信：Liblibaijiang），我们期待您入驻、并将第一时间把模型转移至您的账号中。如您不希望在 LiblibAI 分享该模型，我们也将遵循您的意愿在第一时间下架您的模型。

LiblibAI 尊重每一位模型原创作者，更期待与每一位模型作者共同成长！

声明：若该转载模型引发知识产权纠纷或其他侵权行为，LiblibAI将立即下架模型，并不会向原作者追责。

IronCatLoRA #1 - Artoria Pendragon

Making LoRA's is a thing I do for fun. So if you like what I do? You can help me out by jus* ***ng the LoRA and leaving a review!

This is the first ever LoRA I've attempted to make and after firstly a test training run, than a training run with just auto captioned tags, and then finally having pruned the captioned tags and adding keywords. I feel that this LoRA does the job okay/good enough to be uploaded!

Artoria Pendragon is one of my first favorite characters from Anime, that's why I decided to make my first LoRA about her. So I hope you enjoy this LoRA!

Most images I've uploaded are using the MeinaMix checkpoint, since I like the artsy/anime style a lot from that checkpoint. I've tested it on two other checkpoints I like too; Mistoon_Anime & AbyssOrangeMix2 - SFW . It seems to also kinda work properly on ChilloutMix , I've added one example in the images. I've also uploaded two examples where I also use the 5Shot - Style LoRA , I've found out with experimenting before I made this LoRA that it creates really cool looking fire effects/"atmosphery" etc when combined with MeinaMix.

The main trigger word is artoria pendragon \(fate\) , you can prompt specifically for 3 outfits she wears in the source material, by additionally also using these additional trigger words: armor , dress or suit . Sometimes jus* ***ng suit will not generate her suit properly, using these additional prompts will help fix that: grey shirt , black necktie , black gloves , pant suit & ponytail . You can also prompt for her specific hair styles by using either braid & hair ribbon / ribbon (optionally hair bun ) or ponytail .

The LoRA weights I've used in the images I've uploaded are mostly 0.7 . But this LoRA seems to work okay in a range from 0.7 to 1.0 . The LoRA is also trained on her using Excalibur, although it's currently pretty hard to generate an image where she properly holds Excalibur. You can trigger it via excalibur \(fate/stay night\) . Aditionally using sword and holding sword or hands on hilt or glowing sword might also help, increasing the weight of these prompts might also help. The image I've uploaded where she holds Excalibur proper enough, took me a few generations.

Be sure to also check out my Fate/stay night: UBW Style LoRA , it seems to really work well with this LoRA when using YesMix as checkpoint. The Fate/stay night: UBW Style page includes an image posts, where it shows examples of an images using this LoRA with the Fate/stay night: UBW Style LoRA.

Usage notice: If you copy the prompts from my images here, make sure you change the LoRA part. In some of my prompt you will see Artoria_Pendragon-000019 . But you should change that to, e.g. ironcatlora1Artoria_v10 (default filename when you download from here). Or you should change it to whatever you renamed to file too.

Itseemswithsomecheckpointsthattheadditionaltriggerwordsarmor,dress&suit,sometimesblendintoomuchwithexistingclothingfromtheusedcheckpoint.Butinmytestingitseemsoverallfine,Imightuploadaversion2whereIusededicatedtriggerwordsforher3outfits.

Italsoseemsthathersleevessometimesliketoblendinwhenpromptingotherclothing,thismightbebecausethedatasetIcreatedhasalotofimageswhereshehassleeves.SoIalsomightbetryingtofixthatinanewversion,bytryingtogetmorevarietyinmydataset,inwhattypesofclothingshewears.