Update app.py

app.py

@@ -9,6 +9,7 @@ from typing import List, Optional, Tuple, Union
 from PIL import Image
 import io
 from io import BytesIO
+from diffusers import BitsAndBytesConfig as DiffusersBitsAndBytesConfig
 from diffusers import HunyuanVideoPipeline, FlowMatchEulerDiscreteScheduler
 from diffusers.models.transformers.transformer_hunyuan_video import HunyuanVideoPatchEmbed, HunyuanVideoTransformer3DModel
 from diffusers.utils import export_to_video
@@ -39,7 +40,20 @@ video_transforms = transforms.Compose(
         transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5], inplace=True),
     ]
 )
+quant_config = DiffusersBitsAndBytesConfig(load_in_8bit=True)
+transformer_8bit = HunyuanVideoTransformer3DModel.from_pretrained(
+    "hunyuanvideo-community/HunyuanVideo",
+    subfolder="transformer",
+    quantization_config=quant_config,
+    torch_dtype=torch.bfloat16,
+)
 
+pipeline = HunyuanVideoPipeline.from_pretrained(
+    "hunyuanvideo-community/HunyuanVideo",
+    transformer=transformer_8bit,
+    torch_dtype=torch.float16,
+    device_map="balanced",
+)
 model_id = "hunyuanvideo-community/HunyuanVideo"
 lora_path = hf_hub_download("dashtoon/hunyuan-video-keyframe-control-lora", "i2v.sft")  
 # lora_path = "./cache/i2v.sft"