basic working with llama2

app.py

@@ -1,5 +1,6 @@
 import streamlit as st
 import os
+import time
 from PyPDF2 import PdfReader
 from streamlit_extras.add_vertical_space import add_vertical_space
 from langchain.text_splitter import RecursiveCharacterTextSplitter
@@ -55,14 +56,21 @@ def main():
         "meta/llama-2-70b-chat:02e509c789964a7ea8736978a43525956ef40397be9033abf9fd2badfe68c9e3",
         input={"prompt": prompt}
         )
-        # The predict method returns an iterator, and you can iterate over that output.
-        res=''
-        for item in output:
-            res+=item
-        response = f"AI: {res}"
+        
         # Display assistant response in chat message container
         with st.chat_message("assistant"):
-            st.markdown(response)
+            message_placeholder = st.empty()
+            
+        # The predict method returns an iterator, and you can iterate over that output.
+        response_till_now=''
+        for item in output:
+            response_till_now+=item
+            time.sleep(0.05)
+            message_placeholder.markdown(response_till_now + "▌")
+        message_placeholder.markdown(response_till_now)
+        response = f"AI: {response_till_now}"
+        
+        
         # Add assistant response to chat history
         st.session_state.messages.append({"role": "assistant", "content": response})
             # https://replicate.com/meta/llama-2-70b-chat/versions/02e509c789964a7ea8736978a43525956ef40397be9033abf9fd2badfe68c9e3/api#output-schema