
I made Team Magma Admin, Tabitha, from Pokemon. Idk why he hasn't been made yet even he's pretty fluffy. So, I decide to made LoRA of him instead.
Trigger Word (choose one):
magma_admin_tabitha
Settings I used:
-
Weight: 0.75 - 0.85
-
Sampling Method: DPM++ 2M SDE Karras
-
Sampling steps: 23 - 25
-
Resolution: 768x768 or 960x960 (768x960)
-
If you want to use this following workflow: Generate --> HiRes Fix --> Upscale (Extras) in Automatic1111. Use 640 instead.
-
-
CFG Scale: 7
If you like my LoRA, you can buy me some coffee:
https://www.buymeacoffee.com/doomguy1111
描述:
I updated this model to change it's base checkpoint trained on and the quality of the images. I slashed almost half of my data because the quality of data is low. Here's some changes
-
Decreased image used from 100+ to 60+
-
Increased image quality by removing unecessary background
-
Added proper R34 support. I recommend you to use nsfwEM embedding for maximum result
-
Simplified trigger words
训练词语: magma_admin_tabitha
名称: Tabitha-15(1).safetensors
大小 (KB): 36984
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success