modify

Signed-off-by: MrZ20 <[email protected]>

Author: MrZ20

tests/e2e/multicard/test_aclgraph_capture_replay.py

@@ -134,7 +134,7 @@ def _run_worker_process(
 @pytest.mark.parametrize("model", MODELS)
 @pytest.mark.parametrize("max_tokens", [4, 36])
 @patch.dict(os.environ, {"ASCEND_RT_VISIBLE_DEVICES": "0,1"})
-def test_dp2_aclgraph_capture_replay_metrics(
+def test_aclgraph_capture_replay_metrics_dp2(
     model: str,
     max_tokens: int,
     monkeypatch: pytest.MonkeyPatch,

tests/e2e/multicard/test_data_parallel.py

@@ -38,7 +38,7 @@
 @pytest.mark.parametrize("model", MODELS)
 @pytest.mark.parametrize("max_tokens", [32])
 @patch.dict(os.environ, {"ASCEND_RT_VISIBLE_DEVICES": "0,1"})
-def test_dp_inference_multi_model_configs(model, max_tokens):
+def test_qwen_inference_dp2(model, max_tokens):
     moe_models = ["Qwen/Qwen3-30B-A3B", "vllm-ascend/Qwen3-30B-A3B-W8A8"]
     quantization_models = ["vllm-ascend/Qwen3-30B-A3B-W8A8"]
     script = "examples/offline_data_parallel.py"

tests/e2e/multicard/test_data_parallel_tp2.py

@@ -15,7 +15,7 @@
 @pytest.mark.parametrize("model", MODELS)
 @pytest.mark.parametrize("max_tokens", [32])
 @patch.dict(os.environ, {"ASCEND_RT_VISIBLE_DEVICES": "0,1,2,3"})
-def test_dp2_tp2_hybrid_inference(model, max_tokens):
+def test_qwen_inference_dp2_tp2(model, max_tokens):
     script = "examples/offline_data_parallel.py"
 
     env = os.environ.copy()

tests/e2e/multicard/test_expert_parallel

@@ -5,7 +5,7 @@ from tests.e2e.model_utils import check_outputs_equal
 
 
 @pytest.mark.parametrize("model_name", ["deepseek-ai/DeepSeek-V2-Lite-Chat"])
-def test_ep_correctness_against_tp(model_name):
+def test_deepseek_correctness_ep(model_name):
     example_prompts = [
         "Hello, my name is",
         "The president of the United States is",

tests/e2e/multicard/test_external_launcher.py

@@ -37,7 +37,7 @@
 
 @pytest.mark.parametrize("model", MODELS)
 @patch.dict(os.environ, {"HCCL_BUFFSIZE": "500"})
-def test_external_launcher(model):
+def test_qwen_external_launcher(model):
     script = Path(
         __file__
     ).parent.parent.parent.parent / "examples" / "offline_external_launcher.py"
@@ -78,7 +78,7 @@ def test_external_launcher(model):
 
 
 @pytest.mark.parametrize("model", MOE_MODELS)
-def test_external_launcher_moe_ep_tp2(model):
+def test_qwen_moe_external_launcher_ep(model):
     script = Path(
         __file__
     ).parent.parent.parent.parent / "examples" / "offline_external_launcher.py"
@@ -109,7 +109,7 @@ def test_external_launcher_moe_ep_tp2(model):
 
 
 @patch.dict(os.environ, {"VLLM_ASCEND_ENABLE_NZ": "0"})
-def test_external_launcher_with_sleepmode_level1():
+def test_qwen_external_launcher_with_sleepmode():
     script = Path(
         __file__
     ).parent.parent.parent.parent / "examples" / "offline_external_launcher.py"
@@ -154,7 +154,7 @@ def test_external_launcher_with_sleepmode_level1():
 
 
 @patch.dict(os.environ, {"VLLM_ASCEND_ENABLE_NZ": "0"})
-def test_external_launcher_with_sleepmode_level2():
+def test_qwen_external_launcher_with_sleepmode_level2():
     script = Path(
         __file__
     ).parent.parent.parent.parent / "examples" / "offline_external_launcher.py"
@@ -210,7 +210,7 @@ def test_external_launcher_with_sleepmode_level2():
     "VLLM_ASCEND_ENABLE_MATMUL_ALLREDUCE": "1",
     "HCCL_BUFFSIZE": "500"
 })
-def test_external_launcher_with_matmul_allreduce(model):
+def test_qwen_external_launcher_with_matmul_allreduce(model):
     script = Path(
         __file__
     ).parent.parent.parent.parent / "examples" / "offline_external_launcher.py"

tests/e2e/multicard/test_full_graph_mode.py

@@ -29,7 +29,7 @@
 from tests.e2e.model_utils import check_outputs_equal
 
 
-def test_models_distributed_Qwen3_MOE_TP2_WITH_FULL_DECODE_ONLY():
+def test_qwen_moe_with_full_decode_only():
     if 'HCCL_OP_EXPANSION_MODE' in os.environ:
         del os.environ['HCCL_OP_EXPANSION_MODE']
     prompts = [
@@ -75,7 +75,7 @@ def test_models_distributed_Qwen3_MOE_TP2_WITH_FULL_DECODE_ONLY():
     )
 
 
-def test_models_distributed_Qwen3_MOE_TP2_WITH_FULL():
+def test_qwen_moe_with_full():
     if 'HCCL_OP_EXPANSION_MODE' in os.environ:
         del os.environ['HCCL_OP_EXPANSION_MODE']
     prompts = [

tests/e2e/multicard/test_fused_moe_allgather_ep.py

@@ -41,7 +41,7 @@
         "TASK_QUEUE_ENABLE": "1",
         "VLLM_ENABLE_FUSED_EXPERTS_ALLGATHER_EP": "1"
     })
-def test_generate_with_allgather():
+def test_deepseek_moe_fused_allgather_ep():
     example_prompts = ["Hello, my name is"]
     sampling_params = SamplingParams(max_tokens=100, temperature=0.0)
 
@@ -62,7 +62,7 @@ def test_generate_with_allgather():
     "VLLM_WORKER_MULTIPROC_METHOD": "spawn",
     "TASK_QUEUE_ENABLE": "1"
 })
-def test_generate_with_alltoall():
+def test_deepseek_moe_fused_alltoall_ep():
     example_prompts = ["Hello, my name is"]
     sampling_params = SamplingParams(max_tokens=100, temperature=0.0)
 

tests/e2e/multicard/test_offline_weight_load.py

@@ -31,7 +31,7 @@
 
 @pytest.mark.parametrize("model", MODELS)
 @patch.dict(os.environ, {"VLLM_ASCEND_ENABLE_NZ": "0"})
-def test_offline_weight_load_and_sleepmode(model):
+def test_qwen_offline_weight_load_and_sleepmode(model):
     script = Path(
         __file__
     ).parent.parent.parent.parent / "examples" / "offline_external_launcher.py"

tests/e2e/multicard/test_pipeline_parallel.py

@@ -37,7 +37,7 @@
 @pytest.mark.parametrize("tp_size", TENSOR_PARALLELS)
 @pytest.mark.parametrize("pp_size", PIPELINE_PARALLELS)
 @pytest.mark.parametrize("distributed_executor_backend", DIST_EXECUTOR_BACKEND)
-def test_pp_tp_dist(model: str, tp_size: int, pp_size: int,
+def test_models_pp2(model: str, tp_size: int, pp_size: int,
                     distributed_executor_backend: str) -> None:
     with VllmRunner(model,
                     tensor_parallel_size=tp_size,