v4.3 [final]
I recommend to choose this one or v2. If you don't care about speed and want great anime results use this model + LCM sampler + 20 steps. It's simply better than any model I know in this aspect.
Recommended settings:
6+ steps (8 + 4 steps of upscaler is optimal for my use cases, feel free to experiment).
A few options of samplers:
CFG 1.8 | Euler A
CFG 2.0 | LCM - better for anime and other abstract things.
CFG 1.8 | DPM++ 2M SDE with BETA scheduler (normal doesn't work)
Use style tags such as "anime style, flat coloring" or artist/title name. It will help model to know what you want from it.
The fastest generation: use CFG 1.0 and set lora CFG_slider on negative -0.4 value. Lora listed in suggested resources. (CFG 1.0 ignores negative prompt. Don't use it if you need negative)
The best coherency: use CFG 2.8 and CFG_slider lora on +0.75 weight (for Euler A).
v3.2B
Recommended settings:
CFG 1.8
LCM / [kl optimal / sgm uniform]
5-10 steps
v3.2A
Recommended settings:
CFG 1.8 (works not bad on cfg 1 as well)
Euler A / normal
5-10 steps
v1.99 (v-prediction)
Use title or artist name to get a specific style.
Eauler A - Beta scheduler (or normal)
8-16 steps
CFG: 1.0 recommended for 2x speed. Use more if you want to use negative prompt.
v2
Sampler: Euler A / Euler A cfg++
CFG: (depends on sampler)
Alternative sampler: LCM
Description
I doubt I can do something better than this one by mixing.
FAQ
Comments (4)
Damn you are touching the limits of illustrious capabilities at 8 steps. this insane!
My goal is to save diversity despite using accelerators. For example, you can try to use dmd lora - it will allow you to generate images in 5 steps, but price is too big. Each image will look the same. The most common pose, the most common style, etc... Also prompts with dmd lora are not coherent enough. It looks like I managed to save some coherency and variety by replacing and mixing different blocks of accelerated model.
I tested your new versions of the models a little, and I can say that v4.3 Final is really the best. Now perhaps this model will be my main one, and not v2. Thank you! Your model is personally the best on the site according to my criteria, the others are not just a step behind, but ten. Excellent quality, flexible settings, with phenomenal speed!
I love how fast this model runs! super cool! 💥