debug for qwen2.5 example test in ci

Signed-off-by: yenuo26 <410167048@qq.com>

Author: yenuo26

.buildkite/test-ready.yml

@@ -156,6 +156,7 @@ steps:
         timeout 17m bash -c '
           export VLLM_LOGGING_LEVEL=DEBUG
           export VLLM_WORKER_MULTIPROC_METHOD=spawn
+          pytest -s -v tests/examples/online_serving/test_qwen2_5_omni.py -m "advanced_model" --run-level "advanced_model"
           pytest -s -v tests/e2e/offline_inference/test_qwen2_5_omni.py
         '
     agents:

tests/e2e/stage_configs/qwen2_5_omni_ci.yaml

@@ -22,6 +22,7 @@ stage_args:
       trust_remote_code: true
       engine_output_type: latent
       enable_prefix_caching: false
+      mm_processor_cache_gb: 0
     is_comprehension: true
     final_output: true
     final_output_type: text

tests/examples/online_serving/test_qwen2_5_omni.py

@@ -236,7 +236,6 @@ def test_modality_control_003(omni_server) -> None:
     # TODO: Verify the E2E latency after confirmation baseline.
 
 
-@pytest.mark.skip(reason="There is a known issue with stream error.")
 @pytest.mark.advanced_model
 @pytest.mark.omni
 @hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 4, "rocm": 2})

vllm_omni/model_executor/stage_configs/qwen2_5_omni.yaml

@@ -19,6 +19,7 @@ stage_args:
       engine_output_type: latent
       enable_prefix_caching: false
       max_num_batched_tokens: 32768
+      mm_processor_cache_gb: 0
     is_comprehension: true
     final_output: true
     final_output_type: text

vllm_omni/model_executor/stage_configs/qwen2_5_omni_multiconnector.yaml

@@ -18,6 +18,7 @@ stage_args:
       trust_remote_code: true
       engine_output_type: latent
       enable_prefix_caching: false
+      mm_processor_cache_gb: 0
     is_comprehension: true
     final_output: true
     final_output_type: text