[quantization] AWQ Marlin doesn't work when dtype is bfloat16 (#11494)

Signed-off-by: Kai-Hsun Chen <khchen@x.ai>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>

test/srt/quant/test_awq.py

@@ -1,6 +1,8 @@
 import unittest
 from types import SimpleNamespace
 
+import requests
+
 from sglang.srt.utils import kill_process_tree
 from sglang.test.run_eval import run_eval
 from sglang.test.test_utils import (
@@ -41,5 +43,38 @@ class TestAWQ(CustomTestCase):
         self.assertGreater(metrics["score"], 0.64)
 
 
+class TestAWQMarlinBfloat16(CustomTestCase):
+    """
+    Verify that the model can be loaded with bfloat16 dtype and awq_marlin quantization
+    """
+
+    @classmethod
+    def setUpClass(cls):
+        cls.model = "QuantTrio/Qwen3-VL-30B-A3B-Instruct-AWQ"
+        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=["--dtype", "bfloat16", "--quantization", "awq_marlin"],
+        )
+
+    @classmethod
+    def tearDownClass(cls):
+        kill_process_tree(cls.process.pid)
+
+    def test_mmlu(self):
+        args = SimpleNamespace(
+            base_url=self.base_url,
+            model=self.model,
+            eval_name="mmlu",
+            num_examples=64,
+            num_threads=32,
+        )
+
+        metrics = run_eval(args)
+        self.assertGreater(metrics["score"], 0.88)
+
+
 if __name__ == "__main__":
     unittest.main()