Purpose:
The focus of this style mix is clean, cel-shaded furry generations without the need of heavy prompting or loras, working on your favorite checkpoint. This means smooth colors, expressive characters and clear outlines. Its not meant to be a harsh flat color style but instead more of a softer cel-shaded style, which is I feel is more vibrant and appealing.
The Blend:
Flat Color Style [IL/Pony]
Anhes Lush Puppies Style + Characters
Sigma x Artist Style - [Illustrious]
Jrjresq Artist Style - [Illustrious]
Kaynine Style
Meesh Style
Helpful Tags:
flat color – Pushes the rendering toward stronger cel-shading with flatter colors and minimal shading. This is useful if you want a more traditional animation-style look.
outline – Emphasizes the linework around characters to make them stand out a little more.
Special Thanks and Extra Notes:
A massive thanks to @Piper260 for graciously letting me use their style LoRA. Seriously, half of the LoRAs is in this mix are theirs and for good reason. If you're looking for more cel-shaded styles or furry LoRAs in general, give their profile a look you won't be disappointed.
If you're looking for a checkpoint that can get you consistent cel-shaded art and more without needing this lora, feel free to check out my Indigo Genesis checkpoint. It uses this lora in its mix.
What is an AnthroBlend Style:
AnthroBlend styles are LoRA mixes made by yours truly, aimed at achieving a specific visual style without copying or mimicking any one artist directly. It’s not just a matter of stacking a bunch of similar style LoRAs together and hoping for the best. Each LoRA is individually tuned using block manipulation, which lets me push the parts of the style I want, like shading, line quality, or color behavior, while minimizing things like character design quirks or other stylistic noise baked into the original LoRA. Once the individual pieces are behaving the way I want, they’re combined and tested with carefully chosen weightings. After that, the whole stack is run through a lora optimizer, which adjusts how the LoRAs modify the underlying model weights so they work together instead of stepping on each other’s toes, preventing that muddy look that often happens with multiple LoRAs. The end result is a style that’s clean, consistent, and much easier to use than juggling a pile of separate LoRAs yourself.