Recommended checkpoint: MoyouArtificial and majicmixRealistic

Recommended weight: 0.6-0.8 | <lora:mlnhe:0.8>

Recommended setting: Steps: 28, Sampler: DPM++ SDE Karras, CFG scale: 8

No Trigger word required, but put lora caption first.

Lora trained from Instagram model mlnhe.
Mostly will generate the face and body type, but sometimes slightly generate clothes attire.

Please share your works and leave a review, thank you!