
LoRA based on more or less weird anime images with people and objects coming apart or being segmented; but not in a gory way.
Disclaimer: You probably won't automagically get great results; you need to prompt what you want to see. Flux is your friend, just tell it what to do - or tell ChatGPT to write a prompt for you.
This LoRA is a sibling of Flux Bing Anime Lucid. They are both trained on 80-90s style anime images generated with Dall-E in Bing, so they can do pretty cool/weird things when combined.
All example images use the aiai-Glimra stylizer LoRA
Trigger is aiai-4p4rt, but works without it.
描述:
Optimizer: Prodigy
训练词语: aiai-4p4rt
名称: aiai-4p4rt-Flux-Dev-v6.safetensors
大小 (KB): 167939
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success