diff --git a/diffsynth/models/wan_video_dit.py b/diffsynth/models/wan_video_dit.py
index 43cd601..7386223 100644
--- a/diffsynth/models/wan_video_dit.py
+++ b/diffsynth/models/wan_video_dit.py
@@ -5,7 +5,6 @@ import math
 from typing import Tuple, Optional
 from einops import rearrange
 from .wan_video_camera_controller import SimpleAdapter
-from ..core.device.npu_compatible_device import IS_NPU_AVAILABLE
 
 try:
     import flash_attn_interface
@@ -94,7 +93,7 @@ def rope_apply(x, freqs, num_heads):
     x = rearrange(x, "b s (n d) -> b s n d", n=num_heads)
     x_out = torch.view_as_complex(x.to(torch.float64).reshape(
         x.shape[0], x.shape[1], x.shape[2], -1, 2))
-    freqs = freqs.to(torch.complex64) if IS_NPU_AVAILABLE else freqs
+    freqs = freqs.to(torch.complex64) if freqs.device == "npu" else freqs
     x_out = torch.view_as_real(x_out * freqs).flatten(2)
     return x_out.to(x.dtype)
 
diff --git a/diffsynth/utils/xfuser/xdit_context_parallel.py b/diffsynth/utils/xfuser/xdit_context_parallel.py
index d365cfe..21dc3b3 100644
--- a/diffsynth/utils/xfuser/xdit_context_parallel.py
+++ b/diffsynth/utils/xfuser/xdit_context_parallel.py
@@ -5,7 +5,7 @@ from xfuser.core.distributed import (get_sequence_parallel_rank,
                                      get_sequence_parallel_world_size,
                                      get_sp_group)
 from xfuser.core.long_ctx_attention import xFuserLongContextAttention
-from ...core.device import parse_nccl_backend, parse_device_type, IS_NPU_AVAILABLE
+from ...core.device import parse_nccl_backend, parse_device_type
 
 
 def initialize_usp(device_type):
@@ -50,7 +50,7 @@ def rope_apply(x, freqs, num_heads):
     sp_rank = get_sequence_parallel_rank()
     freqs = pad_freqs(freqs, s_per_rank * sp_size)
     freqs_rank = freqs[(sp_rank * s_per_rank):((sp_rank + 1) * s_per_rank), :, :]
-    freqs_rank = freqs_rank.to(torch.complex64) if IS_NPU_AVAILABLE else freqs_rank
+    freqs_rank = freqs_rank.to(torch.complex64) if freqs_rank.device == "npu" else freqs_rank
     x_out = torch.view_as_real(x_out * freqs_rank).flatten(2)
     return x_out.to(x.dtype)