diff --git a/src/art/local/backend.py b/src/art/local/backend.py
index c9f79e3e..48465ff6 100644
--- a/src/art/local/backend.py
+++ b/src/art/local/backend.py
@@ -315,11 +315,13 @@ async def _monitor_openai_server(
                 if running_requests == 0 and pending_requests == 0:
                     try:
                         # Send a health check with a 5 second timeout
-                        await openai_client.completions.create(
+                        timeout = float(
+                            os.environ.get("ART_SERVER_MONITOR_TIMEOUT", 5.0)
+                        )
+                        # Send a health check with a 5 second timeout
+                        await openai_client.models.retrieve(
                             model=model_name,
-                            prompt="Hi",
-                            max_tokens=1,
-                            timeout=5,
+                            timeout=timeout,
                         )
                     except Exception as e:
                         # If the server is sleeping, a failed health check is okay