LoCon

Any support is appreciated! This will go towards better hardware.https://buymeacoffee.com/tucker_devThis is a LoHA model trained off Reddit. Mostly for teaching myself differen...

https://github.com/deepghs/waifuc/tree/mainThe model data source is derived from Waifuc, a highly robust tool for crawling, filtering, and processing training sets. With the ai...

sakuranomiya maika 桜ノ宮苺香 Blend SUse LyCoris.LoHA with DAdaptAdamWith regularization 描述: V1 训练词语: maika pinkwait,shaded face ...

Support me onfacebook.com/Kaiseirpatreon.com/Serkaihttps://ko-fi.com/kaiseirWeight: 1.0Trigger words:Appearance:raven branwen, long hair, black hair, ponytail, hair ribbon, rib...

Supports all traits, you can check those at Opensea or other marketplaces.cc0_tubbycats is the main triggerword.Traits were shuffled as is with these exceptions:Artist trait wa...

base model: Kohaku-XL Deltaprompts:ffxiv battle talk portrait,simple background, white background, portrait,{you want}neg prompts: Nonesize: 1024*1024two ver different:v1.0 bas...

Train based on AnyLora model. Training materials are basically game CG and character sets.Strongly recommend weights from 0.6 to 0.8,and enable hires.fix.If you find the color ...

Sculpture by Kris Kuksi. A byproduct of testing training parametersThis lora may not be of interest to you, but if you have carried out the training can you leave a comment on ...

Sculpture by Kris Kuksi. A byproduct of testing training parametersThis lora may not be of interest to you, but if you have carried out the training can you leave a comment on ...

Trigger:,theresa \\\(arknights\\\),long hair,horns,dress,white dress,pink hair,long sleeves,bangs,very long hair,red eyes,jewelry,hair between eyes,white hair ...

LyCORIS of Melanie Martinez (samples kindly provided by https://www.tiktok.com/@houstpen_edits - many thanks!)I'm creating Dreambooths, LyCORIS, TIs, and LORAs. Here is my base...

base model: Kohaku-XL Deltaprompts:ffxiv battle talk portrait,simple background, white background, portrait,{you want}neg prompts: Nonesize: 1024*1024two ver different:v1.0 bas...

邦邦人设的三头身小人，记得使用时加上提示词chibi,其实不是很满意，有能力可以自己炼下, 训练素材：与你共享的文件 - OneDrive (live.com) 描述: 训...

recommended weight = 0.8This is a soup-nazi lora I made with kohya-ss. I had seen the seinfeld model and they were saying soup nazi wasn't possible. He was :-).This version I t...

导言这是一个基于Pony模型训练的《Umamusume Pretty Deby》《闪耀优俊少女》的好歌剧角色LyCORIS,使用方法建议以0.8左右的权重使用,在打标过程中，为隔离元素污染，采用了每个卡面分开设关...

Any support is appreciated! This will go towards better hardware.https://buymeacoffee.com/tucker_devThis is a LoHA model trained off Reddit. Mostly for teaching myself differen...

描述: 训练词语: conductor of nephthys,nephthys 名称:...

Rachel McAdams LoHa trained using SD 1.5.Trigger Word: RMcadamsFawziLoHAWeight: 1.0 is the best so far, but test it out for flexibility with different LoRAs.If you prefer using...

https://rentry.co/ProdiAgyThe true iA3. Now within your home... 描述: 训练词语: Mindflayer,teal skin,robot,helmet ...

Works good at any str value, personally i juggle between 1 and 0.8issue: sometimes it generates the car without doorsuse:dsha doors for Dihedral synchro-helix actuation doors (...