
Suggestions
-
Use between 7 and 15 CFG Scale and between 25 and 50 Steps with Heun. Worse samplers might need more steps.
-
To reproduce my results you MIGHT have to change these settings:
-
Set "Do not make Heun deterministic across different batch sizes." (mostly for v1 examples)
-
USE HEUN as sampler
-
Set the ETA Noise Seed Delta (ENSD) to 31337
-
Set CLIP Skip to 2
-
DO not ENABLE face restore for portraits and close ups. It's terrible, messes up with eyes
-
-
Use simple prompts. Complex prompts might make less realistic pictures because of CLIP bleeding. More complex prompts does not mean better results. Keep it simple.
-
Use ADetailer to enhance faces in images having detailed background. Also don't forget to use Restore faces. Basically every solo portrait I made uses it. You can get my settings by clicking on "copy generation data". I suggest you use denoising under 0.3 to avoid getting always the same face.
描述:
I fixed some errors in this version
If you are using loras, you might want to lower CFG
训练词语:
名称: ultimateArt_v20.safetensors
大小 (KB): 5986860
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success