diff --git a/vllm_ascend/spec_decode/eagle_proposer.py b/vllm_ascend/spec_decode/eagle_proposer.py
index 2eef568afe4..0518aa4a57b 100644
--- a/vllm_ascend/spec_decode/eagle_proposer.py
+++ b/vllm_ascend/spec_decode/eagle_proposer.py
@@ -143,7 +143,8 @@ def dummy_run(self,
                   num_tokens_across_dp: Optional[torch.Tensor] = None,
                   aclgraph_runtime_mode: CUDAGraphMode = CUDAGraphMode.NONE,
                   batch_descriptor=None,
-                  dummy_compute_logits=lambda hidden_states: None):
+                  dummy_compute_logits=lambda hidden_states: None,
+                  is_profile=False):
         # update global cos, sin
         update_cos_sin(self.positions[:num_tokens])
 
diff --git a/vllm_ascend/spec_decode/mtp_proposer.py b/vllm_ascend/spec_decode/mtp_proposer.py
index 8bd45bbe5e4..4deef9a2987 100644
--- a/vllm_ascend/spec_decode/mtp_proposer.py
+++ b/vllm_ascend/spec_decode/mtp_proposer.py
@@ -229,7 +229,8 @@ def dummy_run(self,
                   num_tokens_across_dp=None,
                   aclgraph_runtime_mode: CUDAGraphMode = CUDAGraphMode.NONE,
                   batch_descriptor=None,
-                  dummy_compute_logits=lambda hidden_states: None) -> None:
+                  dummy_compute_logits=lambda hidden_states: None,
+                  is_profile=False) -> None:
 
         (
             num_tokens,
@@ -299,7 +300,8 @@ def dummy_run(self,
                     num_actual_tokens=0,
                     aclgraph_runtime_mode=aclgraph_runtime_mode,
                     batch_descriptor=batch_descriptor,
-                    is_mtp_model=True):
+                    is_mtp_model=True,
+                    in_profile_run=is_profile):
                 if self.enable_shared_expert_dp:
                     positions = positions.unsqueeze(-1)
                     positions = torch.ops.vllm.maybe_pad_and_reduce(positions)
diff --git a/vllm_ascend/spec_decode/ngram_proposer.py b/vllm_ascend/spec_decode/ngram_proposer.py
index c5f498c8f87..22d28b61c60 100644
--- a/vllm_ascend/spec_decode/ngram_proposer.py
+++ b/vllm_ascend/spec_decode/ngram_proposer.py
@@ -27,7 +27,8 @@ def dummy_run(self,
                   num_tokens_across_dp=None,
                   aclgraph_runtime_mode: CUDAGraphMode = CUDAGraphMode.NONE,
                   batch_descriptor=None,
-                  dummy_compute_logits=lambda hidden_states: None):
+                  dummy_compute_logits=lambda hidden_states: None,
+                  is_profile=False):
         pass
 
     def generate_token_ids(self,
diff --git a/vllm_ascend/spec_decode/suffix_proposer.py b/vllm_ascend/spec_decode/suffix_proposer.py
index 920b3d4ad2b..ea9f0f72c9d 100644
--- a/vllm_ascend/spec_decode/suffix_proposer.py
+++ b/vllm_ascend/spec_decode/suffix_proposer.py
@@ -27,7 +27,8 @@ def dummy_run(self,
                   num_tokens_across_dp=None,
                   aclgraph_runtime_mode: CUDAGraphMode = CUDAGraphMode.NONE,
                   batch_descriptor=None,
-                  dummy_compute_logits=lambda hidden_states: None):
+                  dummy_compute_logits=lambda hidden_states: None,
+                  is_profile=False):
         pass
 
     def generate_token_ids(self,
diff --git a/vllm_ascend/worker/model_runner_v1.py b/vllm_ascend/worker/model_runner_v1.py
index 4ab4f06d616..b273645ddfe 100644
--- a/vllm_ascend/worker/model_runner_v1.py
+++ b/vllm_ascend/worker/model_runner_v1.py
@@ -2162,7 +2162,8 @@ def dummy_drafter_compute_logits(hidden_states):
                     aclgraph_runtime_mode=aclgraph_runtime_mode,
                     batch_descriptor=batch_descriptor,
                     dummy_compute_logits=dummy_drafter_compute_logits,
-                    in_graph_capturing=not force_attention)
+                    in_graph_capturing=not force_attention,
+                    is_profile=is_profile)
             if is_profile and self.dynamic_eplb:
                 self.model.clear_all_moe_loads()
             if not is_profile and self.dynamic_eplb: