diff --git a/test/srt/models/test_glm4_moe_models.py b/test/srt/models/test_glm4_moe_models.py new file mode 100644 index 000000000..cb742c699 --- /dev/null +++ b/test/srt/models/test_glm4_moe_models.py @@ -0,0 +1,49 @@ +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, + CustomTestCase, + popen_launch_server, +) + + +class TestGLM4MoE(CustomTestCase): + @classmethod + def setUpClass(cls): + cls.model = "zai-org/GLM-4.5-Air-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=[ + "--tp-size", + "2", + ], + ) + + @classmethod + def tearDownClass(cls): + kill_process_tree(cls.process.pid) + + def test_gsm8k(self): + args = SimpleNamespace( + num_shots=5, + data_path=None, + num_questions=100, + 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"], 0.8) + + +if __name__ == "__main__": + unittest.main() diff --git a/test/srt/run_suite.py b/test/srt/run_suite.py index 37eeecdc9..a14d6508c 100644 --- a/test/srt/run_suite.py +++ b/test/srt/run_suite.py @@ -142,6 +142,7 @@ suites = { TestFile("hicache/test_hicache_storage_mooncake_backend.py", 400), TestFile("layers/attention/mamba/test_mamba2_mixer.py", 110), TestFile("lora/test_lora_tp.py", 116), + TestFile("models/test_glm4_moe_models.py", 100), TestFile("rl/test_update_weights_from_distributed.py", 103), TestFile("test_data_parallelism.py", 73), TestFile("test_disaggregation_basic.py", 400),