diff --git a/modelscope/models/cv/video_depth_estimation/dro_model.py b/modelscope/models/cv/video_depth_estimation/dro_model.py
index 761b4b71..ccc0d0a7 100644
--- a/modelscope/models/cv/video_depth_estimation/dro_model.py
+++ b/modelscope/models/cv/video_depth_estimation/dro_model.py
@@ -52,7 +52,8 @@ class DROEstimation(TorchModel):
         if torch.cuda.is_available():
             model_wrapper = self.model_wrapper.to('cuda')
         else:
-            raise RuntimeError('cuda is not available')
+            model_wrapper = self.model_wrapper
+            print('cuda is not available, use cpu')
 
         # Set to eval mode
         model_wrapper.eval()
diff --git a/tests/pipelines/test_video_depth_estimation.py b/tests/pipelines/test_video_depth_estimation.py
index 77cb4b9b..30ca3b33 100644
--- a/tests/pipelines/test_video_depth_estimation.py
+++ b/tests/pipelines/test_video_depth_estimation.py
@@ -15,8 +15,8 @@ class VideoDepthEstimationTest(unittest.TestCase, DemoCompatibilityCheck):
         self.task = 'video-depth-estimation'
         self.model_id = 'damo/cv_dro-resnet18_video-depth-estimation_indoor'
 
-    @unittest.skipUnless(test_level() >= 3, 'skip test in current test level')
-    def test_image_depth_estimation(self):
+    @unittest.skipUnless(test_level() >= 0, 'skip test in current test level')
+    def test_video_depth_estimation(self):
         input_location = 'data/test/videos/video_depth_estimation.mp4'
         estimator = pipeline(Tasks.video_depth_estimation, model=self.model_id)
         result = estimator(input_location)