diff --git a/vllm_ascend/ops/fused_moe/fused_moe.py b/vllm_ascend/ops/fused_moe/fused_moe.py
index f94871fc..12d6a63e 100644
--- a/vllm_ascend/ops/fused_moe/fused_moe.py
+++ b/vllm_ascend/ops/fused_moe/fused_moe.py
@@ -382,9 +382,9 @@ class AscendFusedMoE(FusedMoE):
             group_list_type = fused_experts_results.group_list_type
             assert expert_tokens is not None and group_list_type is not None, \
                 "expert_tokens and group_list_type should not be None when dynamic_eplb is enabled."
-            self.moe_load += expert_tokens if group_list_type == 1 else \
+            local_load = expert_tokens if group_list_type == 1 else \
                 torch.cat([expert_tokens[:1], expert_tokens[1:] - expert_tokens[:-1]])
-
+            self.moe_load.add_(local_load)
         routed_out = forward_context.moe_comm_method.finalize(
             hidden_states=fused_experts_results.routed_out,
             reduce_results=self.reduce_results,