Update handler.py

handler.py

@@ -10,10 +10,10 @@ class EndpointHandler:
         Initialize the model handler using llama_cpp.
         """
         self.model = Llama.from_pretrained(
-            repo_id="bartowski/Llama-3.3-70B-Instruct-GGUF",
-            filename="Llama-3.3-70B-Instruct-IQ4_XS.gguf"
+            repo_id="bartowski/Meta-Llama-3.1-8B-Instruct-GGUF",
+            filename="Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf"
         )
-        self.tokenizer = AutoTokenizer.from_pretrained("bartowski/Llama-3.3-70B-Instruct-GGUF")
+        self.tokenizer = AutoTokenizer.from_pretrained("taylorj94/Llama-3.2-1B")
 
     def get_allowed_token_ids(self, vocab_list: List[str]) -> set[int]:
         """