调整自PuLID ComfyUI，模型全部放到models里的目录，提前下载好，修改依赖，保证安装正常。
PuLID ComfyUI native implementation.
method applies the weights in different ways. Fidelity is closer to the reference ID, Style leaves more freedom to the checkpoint. Sometimes the difference is minimal. I've added neutral that doesn't do any normalization so the reference is very strong and you need to lower the weight.
-
[codeformer-pip]
python_embeded\python.exe -s -m pip install codeformer-pip -
PuLID pre-trained model goes in
ComfyUI/models/pulid/(thanks to Chenlei Hu for converting them into IPAdapter format) -
EVA02_CLIP_L_336_psz14_s6B.pt "ComfyUI\models\pulid\QuanSun\EVA-CLIP\EVA02_CLIP_L_336_psz14_s6B.pt"
-
[parsing_bisenet.pth] "ComfyUI\models\facedetection\parsing_bisenet.pth"
-
[detection_mobilenet0.25_Final.pth] "ComfyUI\models\facedetection\detection_mobilenet0.25_Final.pth"
-
Finally you need InsightFace with AntelopeV2, the unzipped models should be placed in
ComfyUI/models/insightface/models/antelopev2.