Update handler.py

handler.py

@@ -68,6 +68,8 @@ class EndpointHandler():
                                                                       safety_checker=None).to(device)
         # Define Generator with seed
         self.generator = torch.Generator(device="cpu").manual_seed(3)
+        # this command loads the individual model components on GPU on-demand.
+        self.pipe.enable_model_cpu_offload()
 
     def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
         """