阿卡丽 KDA Akali

https://s2.xiaonaofu.top/material/96dcec84-cf91-4234-9253-654fb190d2c1.jpeghttps://s2.xiaonaofu.top/material/a20a253b-c46e-457e-a6f3-37cf2e79c7c4.jpeghttps://s2.xiaonaofu.top/material/d1411039-f2b1-4f53-a4c3-9686e49ee93d.jpeghttps://s2.xiaonaofu.top/material/c1c8af8f-d97e-4cdf-a380-cbb2c6843b08.jpeghttps://s2.xiaonaofu.top/material/f22f5a4e-bbe9-4a6e-90c7-89ae5186f865.jpeghttps://s2.xiaonaofu.top/material/0e6cd710-ce6a-4df6-a839-04590c37496d.jpegkda all out akali

K/DA All Out Akali
Default download is now a LORA

How to use
This is a LORA ckpt which means that you will need https://github.com/kohya-ss/sd-webui-additional-networks to combine the LORA with any model of your choosing.

Generating example images
I’m using AOMV2

Alternative downloads
Alternative downloads were trained via dreambooth

Model Variants
AnythingV3+Protogenx58 mix

Waifu Diffusion V1.3

Creating your own variant
Grab a model from anywhere & go to “Checkpoint Merger” in web-ui. Set A to the downloaded model. Set C to the original model. Set B to the model you want to add it to. Set a custom name, multiplier to 1, select “add difference”. Save as fp16 to save space.

How to create a Dreambooth character model like this!
https://www.reddit.com/r/StableDiffusion/comments/107dw0a/creating_a_specific_character_via_dreambooth/

Hide

发表回复

后才能评论