Doom 3 - Hell Knight

Character LoRA.

My ComfyUI Workflow: Comfyui Workflow: SDXL FaceDetailer, Upscaler, & Batch Selection

Strength: 0.6 to 0.1 is the range. Lots of artistic range. No, "best"

Trigger Word: hellknight, monster

Sampler: Euler_A

Scheduler: Karras

Steps: 20-30

CFG: 6-9. Lots of artistic range, no "right" number.

Tips

Try including both the trigger words hellknight and monster in your prompts

Training tags are included in the files section so you can prompt better outcomes ->

May need to prompt "looking at view" because a lot of the training sets for this LoRA were looking away from the camera.

Training

Base model: WAI-NSFW-illustrious-SDXL. Good for prompting.

All images were tagged using WD14. Touched up colors on a few images to reduce the greenish tint.