
Sharing this realistic portrait model with excellent color and light control, iteratively enhancing realism from my previous ClearMIX version,
The glossy and beautiful realistic photography works I submitted were all made from this model.
Model features: Photography of light and color, clean shape.
I will update the Pro version 1.5 hyper realistic model in the future, which is also the most stable model I have obtained from my research on AI.
Afterwards, I will focus on researching the effectiveness of the XL model.
Recommended usage parameters for the model:
Steps ≥ 30
CFG scale: 8
Size < 1024
My Artstation/Twitter :Art-Dj
分享这个色彩和光感拿捏的很好的写实人像大模型,从我之前的ClearMIX版本迭代强化了写实性,
我投稿的光鲜靓丽的写实摄影作品均是由这个模型制作,
模型特征:光与色彩的摄影,造型干净,
后续将会更新pro版本1.5超写实大模型,也是我研究ai得到的各项数据最稳定的一个大模型,
之后我主力研究xl模型的效果,
模型建议的使用参数:
Steps ≥ 30
CFG scale: 8
Size < 1024
建议放大算法:DPM++ SDE Karras
ART-DJ 设计技术交流摸鱼群:799839872
描述:
Training on a 2.8D model using photos
训练词语:
名称: ClearVAE_V2.3_fp16.safetensors
大小 (KB): 163411
类型: VAE
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: djMix299999_v10.safetensors
大小 (KB): 2246029
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success