
This Lora seems to be able to apply a patch to the generated image, turning an ordinary image into a work of art
You can use this Lora to match various Loras, it can generate better images
I have been following this Fantastic AI Art collection for almost a month now, and the images inside are all very eye-catching. Yesterday, I had a sudden idea: can I use the images inside as a training set to train a LORA? So I selected 567 images from this collection. Unfortunately, after downloading them, I found out that these were just preview images, and their resolutions were only around 450 * 600,
But I still wanted to give it a try, so I packaged these pictures and uploaded them to Civitai. I trained this lora using 521buzz, and the tags were also generated on Civitai. This lora did give me a lot of surprises, just like the name of the Fantastic AI Art collection, Fantastic.
Using ADetailer
I don't know if this is compliant or not. If you think it's not right, contact me to delete this Lora
描述:
训练词语: fantastic ai
名称: 1033369_training_data.zip
大小 (KB): 33291
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: Fantastic_AI.safetensors
大小 (KB): 111774
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success