diff --git a/vllm/vllm-deployment.yaml b/vllm/vllm-deployment.yaml
index 9cb4da7..0237523 100644
--- a/vllm/vllm-deployment.yaml
+++ b/vllm/vllm-deployment.yaml
@@ -13,6 +13,7 @@ spec:
       labels:
         app: vllm-inference-server
     spec:
+      runtimeClassName: nvidia
       containers:
         - name: vllm-inference-server
           image: vllm/vllm-openai:v0.3.3