Upload app.py with huggingface_hub

app.py

@@ -57,7 +57,8 @@ model_kwargs = dict(getattr(config, "model_kwargs", {}) or {})
 efficient_deployment = model_kwargs.pop("efficient_deployment", None) or {}
 # Disable FP8 quantization, lightvae, and custom kernels
 efficient_deployment = {}
-model_kwargs["efficient_deployment"] = efficient_deployment
+# Remove efficient_deployment from model_kwargs to avoid duplicate kwarg
+model_kwargs.pop("efficient_deployment", None)
 
 # Calculate sparse attention sparsity schedule
 num_frame_per_block = getattr(config, "num_frame_per_block", 1)