diff --git a/vllm_ascend/platform.py b/vllm_ascend/platform.py
index 324b693a..0034a653 100644
--- a/vllm_ascend/platform.py
+++ b/vllm_ascend/platform.py
@@ -189,12 +189,6 @@ class NPUPlatform(Platform):
 
         return min(max_num_seqs * decode_query_len, 512)
 
-    @classmethod
-    def apply_config_platform_defaults(cls, vllm_config: VllmConfig) -> None:
-        default_max_cg_capture_size = cls._get_default_max_cudagraph_capture_size(vllm_config)
-        if default_max_cg_capture_size is not None:
-            vllm_config.compilation_config.max_cudagraph_capture_size = default_max_cg_capture_size
-
     @classmethod
     def get_device_capability(cls, device_id: int = 0):
         return None
@@ -209,6 +203,10 @@ class NPUPlatform(Platform):
             pass_config.sp_min_token_num = get_sp_min_token_num(vllm_config)
             logger.info(f"set sp_min_token_num to {pass_config.sp_min_token_num}")
 
+        default_max_cg_capture_size = cls._get_default_max_cudagraph_capture_size(vllm_config)
+        if default_max_cg_capture_size is not None:
+            vllm_config.compilation_config.max_cudagraph_capture_size = default_max_cg_capture_size
+
     @classmethod
     def get_device_name(cls, device_id: int = 0) -> str:
         return torch.npu.get_device_name(device_id)