Anima
Can be used with base model with CFG=4-6, Steps 20-40 or using a Turbo LoRA
Works best with er_sde, euler, res_multistep and heun.
Most example images have "muted colors" and "complex linework" in the prompt and no quality modifiers.
Z-Image
Works with both Z-Image-Base + https://civarchive.com/models/2362961?modelVersionId=2741469 or with Z-Image Turbo.
The LoRA doesn't have a fixed trigger word (was trained with @mikkoani but seems to be irrelevant) but can produce different styles depending on the prompt.
My favorite triggers are:
"anime illustration with muted color and complex linework"
"fashion sketch"
With Z-Image-Base you get more freedom. With Z-Image-Turbo you'll need to be a bit more convincing with your prompt to get the desired effect.
Best used at 1.0, but can be pushed higher, especially with ZIT where at least 1.5 is recommended.
Klein 4B/9B
The LoRA works well both in editing and text-to-image mode. The 4b version can be pushed a little stronger while the 9b version is already quite strong at 1.0. It can give different variations depending on prompt, I like to trigger it with
"@mikkoani illustration with muted colors and complex gritty linework". Adding "anime" also works very well.
Description
New training dataset, generated using the Anima version of this LoRA. Reduces artifacts.
FAQ
Comments (3)
I do not have a GPU so I can only use civitai on-site generation. the configuration here only allows usage of the lora on the z-image base model, not the turbo model. if you just reupload the the model and tag it for turbo I (and others) could use it with turbo then :-)