作者:AIGC101

工作流来源:https://openart.ai/workflows/aigc101/flux-style-transfer-better-version/lJw1nuyXNaGckheGnvqF

load flux ipadapter节点的clip_vision建议使用这个模型:
https://huggingface.co/openai/clip-vit-large-patch14/resolve/main/model.safetensors?download=true
下载完毕后放到.../models/clipvision文件夹下即可
关于controlenet:推荐使用depth的depth anything v2。推荐使用 hed。不推荐canny,使用canny时线条会框定的比较死,如果使用,把权重调低
-------------------------------------
For clip_vision model of "load flux ipadapter" node, I recommends using this model:
https://huggingface.co/openai/clip-vit-large-patch14/resolve/main/model.safetensors?download=true
Just download it and put it in .../models/clipvision folder.
About controlenet: we recommend to use depth anything v2, we recommend to use hed, we don't recommend canny, if you use canny, the line will be framed too much, if you insist, turn down the weights.