Update app.py

app.py

@@ -166,7 +166,7 @@ async def generate(request: Request):
     try:
         response = llm.create_chat_completion(
             messages=messages_for_llm,
-            max_tokens=2048,  # Keep response length short for maximum speed
+            max_tokens=1024,  # Keep response length short for maximum speed
             temperature=0.7, # Adjust temperature for creativity vs. coherence (0.0-1.0)
             stop=["</s>"] # Stop sequence for TinyLlama Chat
         )