How to use ip-adapter-faceid-portrait_sdxl_unnorm.bin

tencent-ailab / IP-Adapter

The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.

Apache License 2.0

4.46k stars 289 forks source link

How to use ip-adapter-faceid-portrait_sdxl_unnorm.bin #386

Open Qsuperme-Q opened 2 weeks ago

Qsuperme-Q commented 2 weeks ago

How do I load the model using diffusers? I have not been able to load the model in the correct way。 from ip_adapter.ip_adapter_faceid_separate import IPAdapterFaceIDXL ip_ckpt = f"{base_path}/IP-Adapter-FaceID/ip-adapter-faceid-portrait_sdxl_unnorm.bin" ip_model = IPAdapterFaceIDXL(pipe, ip_ckpt, device, num_tokens=16, n_cond=5)

Qsuperme-Q commented 2 weeks ago

img_v3_02br_e70916a2-1054-4d84-a3fb-24c5c9aae75g

xiaohu2015 commented 1 week ago

hi, it is only supported at comfyui

Qsuperme-Q commented 1 week ago

hi, it is only supported at comfyui

OK, thanks for your reply