This work flow is set up to produce good results when using the SDXL version of https://tensor.art/models/828264802418930301.
It includes face refining, upscale and upscaled face refining options.
Workflow to use with the Freedive Girls LoRA
Workflow Preview
Showcases (Image/Video)
Showcases (Image/Video)
Nodes Detail
41 NodesPrimitive Node Types
24- PreviewImage6
- SaveImage4
- VAEDecode2
- EmptyLatentImage1
- CLIPSetLastLayer1
- LoraLoader2
- VAEEncode2
- CLIPTextEncode2
- KSampler2
- ImageScaleBy1
- CheckpointLoaderSimple1
Custom Node Types
17- FromBasicPipe4
- NoiseInjectionDetailerHookProvider2
- SAMLoader2
- UltralyticsDetectorProvider2
- MaskToImage2
- ToBasicPipe1
- CR Color Panel1
- FaceDetailer2
- FreeU_V21