About this version
FINALLY, The Chuckie Illustrious LoRA is available. Should work with more models. More examples soon. Use the following tags for best results.
1boy, solo, child, male focus, green shorts, blue shirt, purple-framed eyewear, freckles, buck teeth, orange hair, messy hair, yellow socks, red footwear
In ComfyUI, the strength should be at 1 for show accurate gens. At about 70 or lower, it will confirm to the model. Not every model follows this.
Recommended settings:
Illustrious
This should work with most IL models. More examples soon. Feel free to try out your favorite model and post your examples.
Start your prompt with the following:
1boy, solo, child, male focus, green shorts, blue shirt, purple-framed eyewear, freckles, buck teeth, orange hair, messy hair, yellow socks, red footwear
Danbooru-style tags were used to train this model. I haven't tested it on natural language, but feel free to experiment.
In ComfyUI, the strength_model in the load LoRA node should be at 1 and strength_clip at 2 for show accurate generations. At about .70 or lower, it will confirm to the model. Keep in mind, not every model follows this. My testing so far shows a show accurate generation works best with the base Illustrious model.
Some weirdness
Occasionally, the LoRA generates double pupils in Chuckie's glasses and Saturn on his shirt doesn't come out right, especially with Pony. For Pony, using a different resolution outside of 1024x1024 tends to make him look a bit older. If you want it that way, I recommend doing that, adding "child" in the negative prompt should help. You could also use the tag "1girl"...if you want.