diff --git a/clearml_serving/serving/main.py b/clearml_serving/serving/main.py
index 1896c88..1e3f8fc 100644
--- a/clearml_serving/serving/main.py
+++ b/clearml_serving/serving/main.py
@@ -104,7 +104,7 @@ async def serve_model(model_id: str, version: Optional[str] = None, request: Uni
             instance_id, type(ex), ex, request, "".join(traceback.format_exc())))
         if "CUDA out of memory. " in str(ex) or "NVML_SUCCESS == r INTERNAL ASSERT FAILED" in str(ex):
             # can't always recover from this - prefer to exit the program such that it can be restarted
-            os._exit()
+            os._exit(1)
         raise HTTPException(status_code=422, detail="Error [{}] processing request: {}".format(type(ex), ex))
     except Exception as ex:
         session_logger.report_text("[{}] Exception [{}] {} while processing request: {}\n{}".format(