Update app.py

increased number of output tokens

app.py

@@ -115,9 +115,9 @@ def ask_llm(query, context):
     # 2. Prepare the payload matching your FastAPI 'PromptRequest' schema
     payload = {
         "text": f"USER QUESTION: {query}\n\nDOCUMENT CONTEXT:\n{context[:6000]}",
-        "persona": "You are a Senior Navy Yeoman and Subject Matter Expert. Answer strictly based on the provided context.",
+        "persona": "You are a Senior Navy Yeoman and Subject Matter Expert. Provide a concise answer strictly based on the provided context.",
         "model": "granite4:latest", # You can swap this for 'gemma3:latest' or 'llama3.2:latest' anytime!
-        "max_tokens": 1000
+        "max_tokens": 5000
     }
     
     # 3. Headers for Authentication (Crucial for Private Spaces)
@@ -388,7 +388,9 @@ if query:
             if st.button("✨ Summarize Top Document"):
                 with st.spinner("Reading full document..."):
                     ai_response = ask_llm(query, full_doc_text)
+                    st.markdown("---")
                     st.success(ai_response)
+                    st.markdown("---")
             st.divider()
         # --- LLM INTEGRATION END ---