metadata

license: creativeml-openrail-m
language:
  - en
pipeline_tag: text-to-image
tags:
  - art
  - anime
  - stable-diffusion
library_name: safetensors

∞ 218xl ∞

∞ 218 -- 6 months later

Continuation of what I loved about the first model, with updated ingredients and method of merging, that is Bayesian merging with manual scoring with samples generated with my LoRA.

In addition, I added one of Aha's models into the mix, specifically cV, because I really enjoyed the Counterfeit aesthetics.

And so, this is what I got. I have been using it as my main 1.5 model.

∞ Usage

Amedira cV rules should apply here, so you can make use of "sle" and "mksks style" in the prompt, although I never rigorously tested whether or not they function properly.

I use clip skip 2 and DPM++2M Karras, steps 20

But ultimately whatever you want. I've used 7+ negative embeds, but it's all preference. 0 works just as well.

Yoinked links from Amedira:

Embeddings Downloads

VAE

WD 1.4 Blessed09 (kl-f8-anime2 but blessed down)
WD 1.4 (kl-f8-anime2)

∞ Credits

Checkpoints
Tools
- Bayesian merger by s1dlx

∞ Legal deez

You are free to use this model locally but
1. You aren't allowed to redistribute on another platform. (like CivitAI or Tensor.Art)
2. I am not responsible for how this model is used to generate images.