
UPDATE! I just switched in Civitai the "base model" from SDXL1.0 to Pony because otherwise, as far as i can see, the resource can't be used with Pony model in the generator and tagged with Pony generated pictures (and that's stupid, it can work with both) ^^;
Back on trying to build resources, here is a "Shadowrun" embedding!
It's used in the positive prompt to add element of magic + cyberpunk in the picture.
I built this using embedding-merge extension with the following prompt parsed by words:
<'fantasy'*0.4+'cyberpunk'*1.5+'gothic'*1.1> style, (<'dark'*2.0+'neon'*0.6> vibe:0.6), <'futuristic'*2.0+'magic'*0.7> items, haunting, (dim lighting:1.1), dark shadows, high contrast, cinematic, dramatic, detailed background, (ethereal:0.2)
First try at building an embedding (without going all the way to the dataset) by tricking the CLIP ;-)
描述:
训练词语: sh4d0wrun
名称: sh4d0wrun.safetensors
大小 (KB): 352
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success