diff --git a/vllm_omni/config/endpoint_policy.py b/vllm_omni/config/endpoint_policy.py
index 06731603e..af53b8691 100644
--- a/vllm_omni/config/endpoint_policy.py
+++ b/vllm_omni/config/endpoint_policy.py
@@ -8,7 +8,14 @@ from typing import NamedTuple
 
 from fastapi import FastAPI, Request
 from fastapi.responses import JSONResponse
-from vllm.entrypoints.serve import create_error_response
+
+# vLLM < 0.28 keeps create_error_response under serve.utils (it is not
+# re-exported from vllm.entrypoints.serve); 0.28+ moved it under
+# serve.exception_handling and re-exports it from the package root.
+try:
+    from vllm.entrypoints.serve import create_error_response
+except ImportError:
+    from vllm.entrypoints.serve.utils.error_response import create_error_response
 
 
 class RouteTarget(NamedTuple):
diff --git a/vllm_omni/entrypoints/openai/api_server.py b/vllm_omni/entrypoints/openai/api_server.py
index cee98a3ac..1e55d4476 100755
--- a/vllm_omni/entrypoints/openai/api_server.py
+++ b/vllm_omni/entrypoints/openai/api_server.py
@@ -64,7 +64,14 @@ from vllm.entrypoints.pooling.embed.serving import ServingEmbedding as OpenAISer
 from vllm.entrypoints.pooling.pooling.serving import ServingPooling
 from vllm.entrypoints.pooling.scoring.serving import ServingScores
 from vllm.entrypoints.scale_out.token_in_token_out.serving import ServingTokens
-from vllm.entrypoints.serve import create_error_response
+
+# vLLM < 0.28 keeps create_error_response under serve.utils (it is not
+# re-exported from vllm.entrypoints.serve); 0.28+ moved it under
+# serve.exception_handling and re-exports it from the package root.
+try:
+    from vllm.entrypoints.serve import create_error_response
+except ImportError:
+    from vllm.entrypoints.serve.utils.error_response import create_error_response
 
 # vLLM moved `base` from openai.basic.api_router to serve.instrumentator.basic.
 # Keep a fallback for older/newer upstream layouts during rebase windows.