diff --git a/tests/models/language/generation/test_gemma.py b/tests/models/language/generation/test_gemma.py index ed0f0c19a041..5be4ae874e61 100644 --- a/tests/models/language/generation/test_gemma.py +++ b/tests/models/language/generation/test_gemma.py @@ -7,14 +7,21 @@ @pytest.mark.parametrize("model", MODELS) -def test_dummy_loader(vllm_runner, model: str) -> None: - with vllm_runner( - model, - load_format="dummy", - ) as llm: - normalizers = llm.collective_rpc(lambda self: self.worker.model_runner. - model.model.normalizer.cpu().item()) - assert np.allclose( - normalizers, - llm.llm_engine.model_config.hf_config.hidden_size**0.5, - rtol=1e-3) +def test_dummy_loader(vllm_runner, monkeypatch, model: str) -> None: + with monkeypatch.context() as m: + m.setenv("VLLM_ALLOW_INSECURE_SERIALIZATION", "1") + with vllm_runner( + model, + load_format="dummy", + ) as llm: + if model == "google/gemma-3-4b-it": + normalizers = llm.model.collective_rpc( + lambda self: self.model_runner.model.language_model.model. + normalizer.cpu().item()) + config = llm.model.llm_engine.model_config.hf_config.text_config + else: + normalizers = llm.model.collective_rpc( + lambda self: self.model_runner.model.model.normalizer.cpu( + ).item()) + config = llm.model.llm_engine.model_config.hf_config + assert np.allclose(normalizers, config.hidden_size**0.5, rtol=2e-3)