diff --git a/tests/conftest.py b/tests/conftest.py
index 80e4047c0..143101c44 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -15,6 +15,7 @@ import os
 import signal
 import socket
 import subprocess
+import time
 from typing import Any, Union
 
 import pytest
@@ -70,6 +71,7 @@ class FDRunner:
         if "engine_worker_queue_port" in kwargs:
             ports_to_clean.append(kwargs["engine_worker_queue_port"])
         clean_ports(ports_to_clean)
+        time.sleep(5)
         self.llm = LLM(
             model=model_name_or_path,
             tensor_parallel_size=tensor_parallel_size,
diff --git a/tests/model_loader/test_common_model.py b/tests/model_loader/test_common_model.py
index b7c918411..ef21e52b3 100644
--- a/tests/model_loader/test_common_model.py
+++ b/tests/model_loader/test_common_model.py
@@ -166,7 +166,7 @@ model_param_map = {
         ],
     },
     "Qwen2-7B-Instruct": {
-        "quantizations": ["None", "wint8"],
+        "quantizations": ["wint4"],
     },
     "Qwen3-30B-A3B": {
         "tensor_parallel_size": 2,