Support piecewise cuda graph for Qwen3-next (#13081)
This commit is contained in:
@@ -135,5 +135,43 @@ class TestQwen3NextMTPTopk(CustomTestCase):
|
||||
self.assertGreater(metrics["accuracy"], 0.93)
|
||||
|
||||
|
||||
class TestQwen3NextPiecewiseCudaGraph(CustomTestCase):
|
||||
|
||||
@classmethod
|
||||
def setUpClass(cls):
|
||||
cls.model = "Qwen/Qwen3-Next-80B-A3B-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=[
|
||||
"--tp",
|
||||
"4",
|
||||
"--enable-piecewise-cuda-graph",
|
||||
"--piecewise-cuda-graph-compiler",
|
||||
"eager",
|
||||
],
|
||||
)
|
||||
|
||||
@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(f"{metrics=}")
|
||||
self.assertGreater(metrics["accuracy"], 0.93)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
unittest.main()
|
||||
|
||||
Reference in New Issue
Block a user