Update README.md

README.md

@@ -93,8 +93,10 @@ def retrieve(input_text):
 
 class CallLLM:
     def __init__(self, model_name_or_path: str) -> None:
+        self.quantization_config = BitsAndBytesConfig(load_in_8bit=True)
         self.model = AutoModelForCausalLM.from_pretrained(
             model_name_or_path,
+            quantization_config=self.quantization_config,
             trust_remote_code=True,
             device_map="auto",
         ).eval()