I needed to add "model" param to get it to work

autogenui/manager.py

@@ -21,6 +21,7 @@ def run_flow(self, prompt: str, flow: str = "default") -> None:
             "seed": 42,  # seed for caching and reproducibility
             "config_list": config_list,  # a list of OpenAI API configurations
             "temperature": 0,  # temperature for sampling
+            "model": "gpt-4o",
         }
 
         assistant = autogen.AssistantAgent(
@@ -43,6 +44,7 @@ def run_flow(self, prompt: str, flow: str = "default") -> None:
         user_proxy.initiate_chat(
             assistant,
             message=prompt,
+
         )
 
         messages = user_proxy.chat_messages[assistant]

autogenui/web/app.py

@@ -17,7 +17,7 @@
 # allow cross origin requests for testing on localhost: 800 * ports only
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["http://localhost:8000", "http://localhost:8001", "http://localhost:3000"],
+    allow_origins=["*"],
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],