diff --git a/lightllm/models/gpt_oss/layer_infer/transformer_layer_infer.py b/lightllm/models/gpt_oss/layer_infer/transformer_layer_infer.py
index 4c457fd99..0219495ec 100644
--- a/lightllm/models/gpt_oss/layer_infer/transformer_layer_infer.py
+++ b/lightllm/models/gpt_oss/layer_infer/transformer_layer_infer.py
@@ -65,7 +65,7 @@ def _context_attention_kernel(
         out=None,
     ):
         if self.network_config_["layer_types"][self.layer_num_] == "sliding_attention":
-            window_size = (self.sliding_window - 1, self.sliding_window - 1)
+            window_size = (self.sliding_window - 1, 0)
             use_sliding_window = True
         else:
             window_size = (-1, -1)