This is a merge model curated specifically for LCM. It is optimized for high-quality output at low steps and low CFG scale, with no DMD LORA required.

The model excels at generating character portraits with smooth faces, fine details in the eyebrows, lips, and eyes, and a mature aesthetic. It was built to maximize the quality of the base output to reduce or eliminate the need for aggressive upscaling and face detailer nodes.

Recommended Settings:

• Steps: 10-15

• CFG Scale: 1-2

• Sampler: LCM

• Scheduler: Karras

The models used for this merge are listed in the suggested resources section