Update handler.py

handler.py

@@ -29,10 +29,10 @@ class EndpointHandler():
         
         # Load the models and tokenizer
         self.model = AutoModelForCausalLM.from_pretrained(
-            path,
+            "hypaai/Hypa_Orpheus-3b-0.1-ft-unsloth-merged_16bit",
             torch_dtype=torch.bfloat16
         )
-        self.tokenizer = AutoTokenizer.from_pretrained(path)
+        self.tokenizer = AutoTokenizer.from_pretrained("hypaai/Hypa_Orpheus-3b-0.1-ft-unsloth-merged_16bit")
 
         # Move to devices
         self.device = "cuda" if torch.cuda.is_available() else "cpu"