Update app.py

app.py

@@ -21,7 +21,7 @@ tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
 # Cargar Modelo en AWQ (18-20 GB VRAM)
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
-    torch_dtype=torch.float16,
+    dtype=torch.float16,
     device_map="auto",
     trust_remote_code=True
 )