I adjusted this model to my liking because the colors of the Illustrious model were bland.
Many of the portfolios use <lora:dmd2_sdxl_4step_lora:1>.
Focus of adjustments
1. Relatively clear lines
2. Low brightness colors
3. Presence of color highlights
4. Sharp paintwork
5. Presence of clothing, etc.
Overlooked or weak points
1. Dynamic hypers
2. Tendency to lean toward symmetry
Apologies if there are already many models with the same concept.
While adjusting, I found an anime-style model that I like, so I might upload that one if I feel like it.
Description
example
<lora:dmd2_sdxl_4step_lora:1>Steps: 4, Sampler: LCM, Schedule type: Automatic, CFG scale: 1, VAE: illustriousXLV20_v10.safetensors, Denoising strength: 0.45,Clip skip: 2,Hires CFG Scale: 1, Hires upscale: 1.5,Hires upscaler: Latent (nearest-exact),
There will be a lot of collapse and artifacts. Lora learning and merging