diff --git a/tests/e2e/multicard/test_qwen3_moe.py b/tests/e2e/multicard/test_qwen3_moe.py
index 5dfe36a..a17de55 100644
--- a/tests/e2e/multicard/test_qwen3_moe.py
+++ b/tests/e2e/multicard/test_qwen3_moe.py
@@ -55,6 +55,7 @@ def test_models_distributed_Qwen3_MOE_TP2_WITH_EP():
             tensor_parallel_size=2,
             enable_expert_parallel=True,
             distributed_executor_backend="mp",
+            enforce_eager=False,
     ) as vllm_model:
         vllm_model.generate_greedy(example_prompts, max_tokens)
 
@@ -71,7 +72,7 @@ def test_models_distributed_Qwen3_MOE_W8A8():
             dtype=dtype,
             tensor_parallel_size=2,
             quantization="ascend",
-            enforce_eager=False,
+            enforce_eager=True,
     ) as vllm_model:
         vllm_model.generate_greedy(example_prompts, max_tokens)
 
diff --git a/vllm_ascend/ops/common_fused_moe.py b/vllm_ascend/ops/common_fused_moe.py
index 72ee91b..dd38c23 100644
--- a/vllm_ascend/ops/common_fused_moe.py
+++ b/vllm_ascend/ops/common_fused_moe.py
@@ -242,7 +242,8 @@ class AscendFusedMoE(FusedMoE):
         moe_comm_method_name = forward_context.moe_comm_method_name
 
         # TODO: Can we refactor this logic to model_runner?
-        if not self.moe_config.use_ep:
+        # TODO: Adjusted logic to differentiate between A2 and A3, we check ep_size here since mc2 only support ep_size >= 16 on A3 now
+        if self.moe_config.ep_size < 16:
             moe_comm_method_name = "allgathercommimpl"
 
         forward_context.moe_comm_method = getattr(self, moe_comm_method_name)