Credits to ThatsARepost on civitai!
Trained on bigLust v1.6 Use with DMD2 LoRA, LCM sampler, CFG 1, steps 7-8
What's new in version 4:
Fixed blurry faces, much more higher details overall Applied file size optimization, so LoRA is much smaller despite much higher quality Same training set as v3, but trained at higher rank More flexible, but likeness can sometimes be lower
To create certain type of "art" that has been popular lately follow this steps:
1) Use this: https://tensor.art/models/798218100274141412 on the checkpoint (Big Lust)
2) Use the lora of the celebrity with all the trigger words listed here. (Make sure to not add any negative prompts, they are not needed and it would affect how the images look for worse)
3) Use this lora too: https://tensor.art/models/806617089210062925 (DMD2)
4) Use the following settings: Steps: 7 CFG: 1.0 Sampler: lcm
5)Type and generate whatever you want ;)