diff --git a/test/srt/models/test_qwen_models.py b/test/srt/models/test_qwen_models.py
new file mode 100644
index 00000000000..903fd45d550
--- /dev/null
+++ b/test/srt/models/test_qwen_models.py
@@ -0,0 +1,78 @@
+import unittest
+from types import SimpleNamespace
+
+from sglang.srt.utils import kill_process_tree
+from sglang.test.few_shot_gsm8k import run_eval
+from sglang.test.test_utils import (
+    DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+    DEFAULT_URL_FOR_TEST,
+    popen_launch_server,
+)
+
+
+class TestQwen2(unittest.TestCase):
+    @classmethod
+    def setUpClass(cls):
+        cls.model = "Qwen/Qwen2-7B-Instruct"
+        cls.base_url = DEFAULT_URL_FOR_TEST
+        cls.process = popen_launch_server(
+            cls.model,
+            cls.base_url,
+            timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+            other_args=[],
+        )
+
+    @classmethod
+    def tearDownClass(cls):
+        kill_process_tree(cls.process.pid)
+
+    def test_gsm8k(self):
+        args = SimpleNamespace(
+            num_shots=5,
+            data_path=None,
+            num_questions=200,
+            max_new_tokens=512,
+            parallel=128,
+            host="http://127.0.0.1",
+            port=int(self.base_url.split(":")[-1]),
+        )
+        metrics = run_eval(args)
+        print(metrics)
+
+        self.assertGreater(metrics["accuracy"], 0.81)
+
+
+class TestQwen2FP8(unittest.TestCase):
+    @classmethod
+    def setUpClass(cls):
+        cls.model = "neuralmagic/Qwen2-7B-Instruct-FP8"
+        cls.base_url = DEFAULT_URL_FOR_TEST
+        cls.process = popen_launch_server(
+            cls.model,
+            cls.base_url,
+            timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+            other_args=[],
+        )
+
+    @classmethod
+    def tearDownClass(cls):
+        kill_process_tree(cls.process.pid)
+
+    def test_gsm8k(self):
+        args = SimpleNamespace(
+            num_shots=5,
+            data_path=None,
+            num_questions=200,
+            max_new_tokens=512,
+            parallel=128,
+            host="http://127.0.0.1",
+            port=int(self.base_url.split(":")[-1]),
+        )
+        metrics = run_eval(args)
+        print(metrics)
+
+        self.assertGreater(metrics["accuracy"], 0.8)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/test/srt/run_suite.py b/test/srt/run_suite.py
index 83e24e3a867..e2ecdfb3a68 100644
--- a/test/srt/run_suite.py
+++ b/test/srt/run_suite.py
@@ -8,6 +8,7 @@
         "models/test_embedding_models.py",
         "models/test_generation_models.py",
         "models/test_lora.py",
+        "models/test_qwen_models.py",
         "models/test_reward_models.py",
         "sampling/penaltylib",
         "test_abort.py",