Update handler.py

handler.py

@@ -11,7 +11,7 @@ class EndpointHandler:
     def __init__(self, path=""):
         # load the model
         tokenizer = AutoTokenizer.from_pretrained(path)        
-        model = AutoModelForCausalLM.from_pretrained(path, device_map="auto", torch_dtype=dtype, trust_remote_code=True)
+        model = AutoModelForCausalLM.from_pretrained(path, load_in_8bit=True, device_map="auto", torch_dtype=dtype, trust_remote_code=True)
     
         # create inference pipeline
         self.pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer, device='cuda:0')