diff --git a/swift/megatron/init.py b/swift/megatron/init.py
index 7f64559423..7b7ebc0f3e 100644
--- a/swift/megatron/init.py
+++ b/swift/megatron/init.py
@@ -745,12 +745,7 @@ def _apply_rotary_pos_emb_thd(
         Returns:
             Tensor: Shape [t, h, d]. The input tensor after applying RoPE.
         """
-        if cp_group is not None:
-            cp_size = cp_group.size()
-        else:
-            cp_size = mpu.get_context_parallel_world_size()
-        cu_seqlens_for_batched = cu_seqlens // cp_size
-        use_batched_rope = (freqs.dim() >= 1 and freqs.shape[0] == cu_seqlens_for_batched[-1]).item()
+        use_batched_rope = freqs.dim() >= 1 and freqs.shape[0] == t.shape[0]
         if not use_batched_rope:
             logger.warning_once('Using non-batched RoPE, which may affect performance.')
             kwargs = {'cp_group': cp_group} if mcore_013 else {}