diff --git a/tests/e2e/offline_inference/test_qwen3_tts_customvoice.py b/tests/e2e/offline_inference/test_qwen3_tts_customvoice.py
index 416a5009c..4fa2d4b6f 100644
--- a/tests/e2e/offline_inference/test_qwen3_tts_customvoice.py
+++ b/tests/e2e/offline_inference/test_qwen3_tts_customvoice.py
@@ -34,12 +34,10 @@ def get_cuda_graph_config():
             "stages": {
                 0: {
                     "max_num_seqs": 1,
-                    "gpu_memory_utilization": 0.2,
                     "enforce_eager": True,
                     "async_scheduling": False,
                 },
                 1: {
-                    "gpu_memory_utilization": 0.2,
                     "enforce_eager": True,
                     "async_scheduling": False,
                 },
diff --git a/tests/model_executor/stage_input_processors/test_qwen3_omni_streaming_helpers.py b/tests/model_executor/stage_input_processors/test_qwen3_omni_streaming_helpers.py
index afc71c476..811620608 100644
--- a/tests/model_executor/stage_input_processors/test_qwen3_omni_streaming_helpers.py
+++ b/tests/model_executor/stage_input_processors/test_qwen3_omni_streaming_helpers.py
@@ -247,7 +247,6 @@ def test_thinker2talker_full_payload_packs_complete_tensors() -> None:
     assert payload["ids"]["all"] == [151644, 872, 3]
     assert payload["embed"]["prefill"].device.type == "cpu"
     assert payload["hidden_states"]["output"].device.type == "cpu"
-    assert payload["meta"]["next_stage_prompt_len"] > 0
     assert payload["embed"]["prefill"].shape[0] == 2
     assert payload["hidden_states"]["output"].shape[0] == 2