22

app.py

@@ -14,14 +14,15 @@ from image_utils import get_image_latent, scale_image
 # from videox_fun.utils.utils import get_image_latent
 
 
-MODEL_REPO = "Tongyi-MAI/Z-Image-Turbo"
+# MODEL_REPO = "Tongyi-MAI/Z-Image-Turbo"
 MAX_SEED = np.iinfo(np.int32).max
 MAX_IMAGE_SIZE = 1280
 
+# git clone https://huggingface.co/Tongyi-MAI/Z-Image-Turbo
 MODEL_LOCAL = "models/Z-Image-Turbo/"
+# curl -L -o Z-Image-Turbo-Fun-Controlnet-Union.safetensors https://huggingface.co/alibaba-pai/Z-Image-Turbo-Fun-Controlnet-Union/resolve/main/Z-Image-Turbo-Fun-Controlnet-Union.safetensors
 TRANSFORMER_LOCAL = "models/Z-Image-Turbo-Fun-Controlnet-Union.safetensors"
 
-
 weight_dtype = torch.bfloat16
 
 # load transformer
@@ -48,8 +49,7 @@ if TRANSFORMER_LOCAL is not None:
     m, u = transformer.load_state_dict(state_dict, strict=False)
     print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
-# Load MODEL_REPO
-# Get Vae
+# load ZImageControlPipeline
 vae = AutoencoderKL.from_pretrained(
     MODEL_LOCAL,
     subfolder="vae"