diff --git a/interpreter/cli.py b/interpreter/cli.py
index 5b81e4622a..0d38d526e6 100644
--- a/interpreter/cli.py
+++ b/interpreter/cli.py
@@ -186,12 +186,6 @@ async def async_load_interpreter(args):
 async def async_main(args):
     global global_interpreter
 
-    if args["serve"]:
-        global_interpreter = await async_load_interpreter(args)
-        print("Starting server...")
-        global_interpreter.server()
-        return
-
     if (
         args["input"] is None
         and sys.stdin.isatty()
diff --git a/interpreter/interpreter.py b/interpreter/interpreter.py
index 29c93e7abc..695b2e80a9 100644
--- a/interpreter/interpreter.py
+++ b/interpreter/interpreter.py
@@ -1073,6 +1073,21 @@ def server(self):
         # Create and start server
         server = Server(self)
         try:
+            host = server.host
+            port = server.port
+
+            print("\n" + "=" * 60)
+            print(f"Open Interpreter API Server")
+            print("=" * 60)
+            print("\nTo use with an OpenAI-compatible client, configure:")
+            print(f"  - API Base:     http://{host}:{port}")
+            print(f"  - API Path:     /chat/completions")
+            print(f"  - API Key:      (any value, authentication not required)")
+            print(f"  - Model name:   (any value, ignored)")
+            print("\nNOTE: The server will use the model configured in --model")
+            print(f"      Currently using: {self.model}")
+            print("=" * 60 + "\n")
+
             server.run()
         except KeyboardInterrupt:
             print("\nShutting down server...")