
This LoRA is an effort to bring Metal Slug to XL, and vastly improve on the quality of generations.
You have a high amount of variety with this LoRA as it includes training data from the classics, MS Attack Reloaded, MS Tactics, mobile games and promotional artwork:
[Triggers]
• mspixel, metalslugxl, pixelated, short stack, full body, slim
• illustration, metalslugxl, cowboy shot
• concept art, metalslugxl, reference sheet
• 3D, metalslugxl, render
[Characters}
• marco, tarma, fio, eri, trevor, ralph, clark, leona, nardia, general morden, rita, yoshino, metzelei, dragonuv
[Settings]
Sampler: DPM++2M
Steps: 38
Res: 1024 x 1024
Model: Pony
For pixel perfect results, reduce to 7% native resolution with nearest neighbour scaling. Touch up any minor artifacts using the pencil tool.
[Compatibility]
Some overtuned LoRA's may overpower the style. You can fix this by turning down the weight of your custom character LoRA in img-to-img to re-apply the pixelation effect.
描述:
训练词语: metalslugxl,mspixel,pixelated,short stack,slim,short legs,full body,from side
名称: metalslugxl.safetensors
大小 (KB): 223144
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success