diff --git a/tests/integration/defs/accuracy/test_llm_api_autodeploy.py b/tests/integration/defs/accuracy/test_llm_api_autodeploy.py index 632ee879409b..5b014a689b31 100644 --- a/tests/integration/defs/accuracy/test_llm_api_autodeploy.py +++ b/tests/integration/defs/accuracy/test_llm_api_autodeploy.py @@ -498,7 +498,7 @@ def evaluate_tasks(self, llm, sampling_params): task = GSM8K(self.MODEL_NAME) task.evaluate(llm) - @pytest.mark.skip_less_device_memory(32000) + @pytest.mark.skip_less_device_memory(80000) @pytest.mark.parametrize("enable_chunked_prefill", [True, False]) def test_auto_dtype(self, enable_chunked_prefill): kwargs = self.get_default_kwargs(enable_chunked_prefill)