diff --git a/tests/entrypoints/openai/test_default_mm_loras.py b/tests/entrypoints/openai/test_default_mm_loras.py index 372e9b1fecd4..b9c466a6fbeb 100644 --- a/tests/entrypoints/openai/test_default_mm_loras.py +++ b/tests/entrypoints/openai/test_default_mm_loras.py @@ -48,7 +48,8 @@ def multimodal_server(): # noqa: F811 f"{{\"audio\": \"{AUDIO_LORA_PATH}\"}}", ] - with RemoteOpenAIServer(MULTIMODAL_MODEL_NAME, args) as remote_server: + with RemoteOpenAIServer(MULTIMODAL_MODEL_NAME, args, + max_wait_seconds=480) as remote_server: yield remote_server diff --git a/tests/mq_llm_engine/test_error_handling.py b/tests/mq_llm_engine/test_error_handling.py index 3feee01dadf7..77e3732cd06c 100644 --- a/tests/mq_llm_engine/test_error_handling.py +++ b/tests/mq_llm_engine/test_error_handling.py @@ -255,8 +255,8 @@ async def test_mp_crash_detection(monkeypatch: pytest.MonkeyPatch): pass end = time.perf_counter() - assert end - start < 60, ( - "Expected vLLM to gracefully shutdown in <60s " + assert end - start < 100, ( + "Expected vLLM to gracefully shutdown in <100s " "if there is an error in the startup.")