Update app.py

app.py

@@ -16,7 +16,7 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 if torch.cuda.is_available():
     model_id = "Qwen/Qwen2.5-7B-Instruct"
     #model_id = "BenBranyon/sumbot7b-augmented"
-    model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto")
+    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
     tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True, model_max_length=MAX_INPUT_TOKEN_LENGTH, padding="longest", language="en")
 
 #Transformers Code