Update app.py

app.py

@@ -176,7 +176,7 @@ def preprocess_text_with_llm(input_text: str) -> str:
         response = ""
         stream = hf_client.text_generation(
             prompt=system_prompt,
-            max_new_tokens=4000,
+            max_new_tokens=2000,
             temperature=0.5,
             top_p=0.9,
             stream=True,