
Alphonse Mucha's style is so distinct and I absolutely adore it. This lora was trained on 170 of his works, which are all public domain. I used the highest quality of each image I could find; unfortunately, that was still pretty low res for some images, but I still had pretty good results with the sharpness on generated images. It works very well with character loras as well!
You can vary the strength of the style without changing the weight of the lora itself by changing the triggers in your prompt:
-
strongest style:
mucha_style, a painting by Alphonse Mucha depicting ...
-
slightly less strong but still strong style:
a painting by Alphonse Mucha depicting ...
-
some style elements but not full-blown Mucha style:
mucha_style, a painting depicting ...
See the gallery for examples of the differences in these prompts!
As always, I would love to hear your thoughts and see your images. ?
描述:
训练词语: mucha_style,a painting by Alphonse Mucha
名称: 771940_training_data.zip
大小 (KB): 45662
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: Alphonse_Mucha-000008.safetensors
大小 (KB): 18975
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success