Update app.py

app.py

@@ -471,7 +471,7 @@ def get_response_from_excel(query, model, context, num_calls=3, temperature=0.2)
 
         logging.info("Finished generating response for Excel data")
 
-def truncate_context(context, max_chars=24000):
+def truncate_context(context, max_chars=20000):
     """Truncate context to a maximum number of characters."""
     if len(context) <= max_chars:
         return context
@@ -537,7 +537,7 @@ def get_response_from_llama(query, model, selected_docs, file_type, num_calls=1,
             # Generate content with streaming enabled
             for response in client.chat_completion(
                 messages=messages,  # Pass messages in the required format
-                max_tokens=20000,  # Reduced to ensure we stay within token limits
+                max_tokens=3000,  # Reduced to ensure we stay within token limits
                 temperature=temperature,
                 stream=True,
                 top_p=0.9,