modify max tokens

app.py

@@ -131,7 +131,7 @@ class GradioApp:
             max_token = gr.Slider(
                 128,
                 1024,
-                value=256,
+                value=128,
                 step=1,
                 label=self._locale("Max tokens"),
                 info=self._locale("The maximum number of tokens to generate."),
@@ -462,7 +462,7 @@ async def launch_xinference():
         address=supervisor_address, supervisor_address=supervisor_address
     )
     api = AsyncSupervisorAPI(supervisor_address)
-    supported_models = ["orca", "chatglm2", "chatglm", "vicuna-v1.3"]
+    supported_models = ["chatglm2", "chatglm", "vicuna-v1.3", "orca"]
     for model in supported_models:
         await api.launch_model(str(uuid.uuid4()), model)