Update main.py

main.py

@@ -1,12 +1,20 @@
 import SpaceFace as sf
 import GPTSimple as ai
+import time
 class AIEndpoint:
     def receivedMessage(self, message, user: sf.User, responder: sf.Responder):
         if hasattr(user, 'chat'):
             
-            responder.respond(user.chat.ask(message).text)
+            answer = user.chat.ask(message, stream=True)
         else:
             user.chat = ai.conversation(base_url="deepinfra", model="meta-llama/Llama-3.2-90B-Vision-Instruct")
-            responder.respond(user.chat.ask(message).text)
+            answer = user.chat.ask(message, stream=True)
+        r=""
+        for token in answer:
+            if token:
+                r+=token.token.replace("\n", "\\n")
+                if time.time() - self.cooldown >= 0.5:
+                    responder.respond(r)
+                    r=""
         responder.close()
 sf.start_server([AIEndpoint()])