:closed_lock_with_key: Update authentication to inference server

app.py

@@ -14,6 +14,7 @@ STOP_STR = f"\n{USER_NAME}:"
 STOP_SUSPECT_LIST = [":", "\n", "User"]
 
 INFERENCE_ENDPOINT = os.environ.get("INFERENCE_ENDPOINT")
+INFERENCE_AUTH = os.environ.get("INFERENCE_AUTH")
 
 
 def chat_accordion():
@@ -214,7 +215,7 @@ if __name__ == "__main__":
         default=INFERENCE_ENDPOINT,
     )
     args = parser.parse_args()
-    client = Client(args.addr)
+    client = Client(args.addr, headers={"Authorization": f"Basic {INFERENCE_AUTH}"})
     demo = get_demo(client)
     demo.queue(concurrency_count=2)
     demo.launch()