Most of the sample pictures on all loras are done with controlnet/img2img so expect different results if you trying to remix with the civitai generator.
For local users this is the model of controlnet I use.
https://huggingface.co/2vXpSwA7/iroiro-lora/blob/main/test_controlnet2/CN-anytest_v4-marged.safetensors
I use it at 0.5 strength without any preprocessor.
30/09/24
Edited somethingweird lora page to use as a repo for AI artstyles, might need to re select it on the generator.
描述:
30/09/24
Moved from their own lora page to this one.
Lora based on SD 1.5 images made by me.
The dataset for this was just brute forced from 6k images so it will be best to use at a lower strength 0.7 or lower.
训练词语:
名称: [Artstyle] Thicc_Geekpower [PDXL].safetensors
大小 (KB): 166577
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
