Update app.py

app.py

@@ -9,8 +9,8 @@ prompt_ar = "### Instruction: اسمك جيس وسميت على اسم جبل ج
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
 tokenizer = AutoTokenizer.from_pretrained(model_path)
-model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto", trust_remote_code=True)
-
+#model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(model_path, torch_dtype=torch.float16)
 
 def get_response(text,tokenizer=tokenizer,model=model):
     input_ids = tokenizer(text, return_tensors="pt").input_ids