fix: enable multi-threading for h200 tests (#16413)

2026-01-04 14:43:53 -08:00
parent a1ed247fd7
commit 55d112dc79
6 changed files with 23 additions and 1 deletions
@@ -174,6 +174,8 @@ class TestDeepseekV32TBO(CustomTestCase):
            "deepep",
            "--cuda-graph-max-bs",
            "256",
+            "--model-loader-extra-config",
+            '{"enable_multithread_load": true, "num_threads": 64}',
        ]
        cls.process = popen_launch_server(
            cls.model,