Sometimes I like to take a break from the search for realism, ever since I discovered Illustrious I've wanted to make a base model in the ComiX style that I like, what is this model capable of? Creating basic images with thick lines that emphasize the comic style and creating slightly more stylized images, I put a lot of emphasis on understanding the prompt as well as facial expressions, I hope you enjoy it as much as I do and share your examples.
The VAE is baked, That's right, once again
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 15 denois 0.35 Hires CFG 4,
Width and height: 896*1152
CFG Scale: 5
Steps: 35
The sampling method i use is Euler A-Karras
Neg prompt: worst quality, low quality
Description
The changes and the main focus of this model are on improving facial expressions, thick lines emphasizing the comic style and in general a more toon aesthetic.
The VAE is baked, That's right, once again
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 15 denois 0.35 Hires CFG 4,
Width and height: 896*1152
CFG Scale: 5
Steps: 35
The sampling method i use is Euler A-Karras
Neg prompt: worst quality, low quality
FAQ
Comments (4)
v3 can easily overdo it with the thick black character outline, unless that was your intention?
Post with thick lines: https://civitai.com/posts/20918718
Post with "(no lineart:1.5),": https://civitai.com/posts/20918794
That was the intention, i love the aestethics of the thick outlines, I don't think it's necessarily an improvement, it's just different from before ;)
Sateluco The lora you used for the facial expressions (niji? expressiveh?) leans heavily towards the upturned, eyes half-closed look. To avoid need to prompt for eyes. Nicely different from v2.