Update app.py

app.py

@@ -49,8 +49,10 @@ def generate(prompt, system, history, temperature=0.9, max_new_tokens=4096, top_
             response_text = response.split("ASSISTANT:")[-1].strip()
 
             # Simulate streaming by yielding parts of the response
+            accumulated_response = ""  # To keep track of the full response
             for char in response_text:
-                yield char
+                accumulated_response += char  # Append the new character
+                yield accumulated_response  # Yield the accumulated response
                 time.sleep(0.05)  # Add a slight delay to simulate typing
 
             # Save the generated response to the database after the response is generated