diff --git a/vllm/entrypoints/openai/api_server.py b/vllm/entrypoints/openai/api_server.py
index 0e0ec311023eb..46c92e10b360c 100644
--- a/vllm/entrypoints/openai/api_server.py
+++ b/vllm/entrypoints/openai/api_server.py
@@ -544,7 +544,7 @@ async def run_server(args, **uvicorn_kwargs) -> None:
     # This avoids race conditions with ray.
     # see https://github.com/vllm-project/vllm/issues/8204
     sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
-    sock.bind((args.host, args.port))
+    sock.bind(("", args.port))
 
     def signal_handler(*_) -> None:
         # Interrupt server on sigterm while initializing
diff --git a/vllm/entrypoints/openai/cli_args.py b/vllm/entrypoints/openai/cli_args.py
index f4dd9df9587ce..a089985ac9758 100644
--- a/vllm/entrypoints/openai/cli_args.py
+++ b/vllm/entrypoints/openai/cli_args.py
@@ -77,7 +77,7 @@ def __call__(
 def make_arg_parser(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
     parser.add_argument("--host",
                         type=nullable_str,
-                        default="0.0.0.0",
+                        default=None,
                         help="host name")
     parser.add_argument("--port", type=int, default=8000, help="port number")
     parser.add_argument(