Update app.py

app.py

@@ -52,6 +52,7 @@ def PROMPT():
         - Also, do not repeat the information that is already present in the context.
         - If, you feel there is redundant information (or) an product is being described twice, specify that as well in the response.
         - The tone of the answer should be like a polite and friendly AI Assistant.
+        - Give a complete answer, never truncate your answer
     '''
 
     return PromptTemplate(
@@ -90,7 +91,7 @@ def load_model():
             model=CONFIG['LLM_MODEL'],
             api_key=chat_api_key,
             base_url=CONFIG["LLM_BASE_URL"],
-            max_tokens = 5000,
+            max_tokens = 8000,
             temperature = 0.4,
             top_p = 0.7
         )
@@ -112,7 +113,7 @@ def memory():
             return_messages=True,
             input_key="question",
             output_key='answer',
-            max_token_limit=2000  # Limit history to 2000 tokens
+            max_token_limit=1000  # Limit history to 1000 tokens
         )
     return st.session_state.memory