managing token limit

app.py

@@ -132,7 +132,7 @@ def on_submit(symptoms_text, history):
     yield history, None, "\n".join(log)
 
     # Call LLM
-    response = llm.complete(prompt)
+    response = llm(prompt=prompt)
     raw = response
     # Extract text from CompletionResponse if needed
     if hasattr(raw, "text"):