diff --git a/docker/llm/serving/xpu/docker/README.md b/docker/llm/serving/xpu/docker/README.md
index fd5826fee35..9109707c3a9 100644
--- a/docker/llm/serving/xpu/docker/README.md
+++ b/docker/llm/serving/xpu/docker/README.md
@@ -102,7 +102,9 @@ To set up model serving using `IPEX-LLM` as backend using FastChat, you can refe
     
     # start controller
     python -m fastchat.serve.controller &
-    
+
+    export USE_XETLA=OFF
+    export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=2
     
     export TORCH_LLM_ALLREDUCE=0
     export CCL_DG2_ALLREDUCE=1
diff --git a/docs/mddocs/DockerGuides/vllm_docker_quickstart.md b/docs/mddocs/DockerGuides/vllm_docker_quickstart.md
index 91ef0d34c70..5396ae242b9 100644
--- a/docs/mddocs/DockerGuides/vllm_docker_quickstart.md
+++ b/docs/mddocs/DockerGuides/vllm_docker_quickstart.md
@@ -831,7 +831,9 @@ We can set up model serving using `IPEX-LLM` as backend using FastChat, the foll
     
     # start controller
     python -m fastchat.serve.controller &
-    
+
+    export USE_XETLA=OFF
+    export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=2
     
     export TORCH_LLM_ALLREDUCE=0
     export CCL_DG2_ALLREDUCE=1