This is my first mix model. I'm not really sure what the best practices are; I just picked a bunch of models I found to work well together and merged them with varying weights.
It aims for a very peculiar style that's sort of flat/painterly, so if you aren't looking for that specific sort of style in your images all the time you might not like using it.
That being said it can produce some nice character images. Sometimes the outlines have a fuzzy "digital chalk/crayon"-kind of effect I was wanting.
Also note: there are no guarantees that all LoRAs will respond well with this merge. I think it should work okay with most concept LoRAs though. I recommend using kakigori v3 instead (the original model I merged into) or its own source models if you want a less opinionated model to try.
Base model: https://civitai.com/models/100505/kakigori (V3)
Recommended tags: surreal, abstract, complimentary colors
Also try: traditional media, negative space, collage, vincent van gogh
Negative embeddings tested: EasyNegative, negative_hand-neg, BadDream, UnrealisticDream
I usually use Euler_a with CFG Scale 6-8. Also CFG Scale 3-4 sometimes gives good results depending on the prompt.
I'm happy using the original NovelAI VAE with this model.
You can see the example images for positive/negative prompt ideas. A typical positive/negative prompt is like kakigori's.
Positive:
masterpiece, surreal, abstract, 1girl, dress, floral print, long hair, straight hair
Negative:
(worst quality, low quality:1.2), border, nsfw, skimpy, grayscale, multiple girls, (watermark:1.2)
NOTE: I use ComfyUI so the examples will probably not replicate exactly in WebUI, but as far as I can tell they come pretty close.
The following text is copied from kakigori's license:
This model may not be used on any paid generation services. You may not sell this model or any merges that contain it. This model is for personal/private use. Commercial use is not authorized.
You are more than welcome to include this model in merges as long as those merges adhere to the same license as this model. If you upload your merge all I ask is that you include a link to this model in the about section like I have done here.