
Ausum LoRA enhances images by subtly introducing more exciting details and effects, enriching the visuals without overpowering the original composition. The idea was to design a lora to elevate the aesthetic while maintaining the integrity of the base image.
While i am a bit far from my original goal, the lora is in a good place to be released, i thought i take notes of the community to help me improve.
Use between 0.5 ~ 1.5. You can go higher but i do not recommend it.
Any note or advice is welcome.
Thanks and enjoy.
描述:
Trained on 410 images.
You can make the Lora more aggressive by increasing weight, bot no more than 2.
Any image you post can help me acquire more knowledge towards enhancing the lora and do a better job for the flux and pony version of it.
训练词语: ausum style
名称: Ausum-SDXL-V1.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