diff --git a/norch/nn/module.py b/norch/nn/module.py
index cebd86f..eaabd9a 100644
--- a/norch/nn/module.py
+++ b/norch/nn/module.py
@@ -53,6 +53,8 @@ class Module(ABC):
     def to(self, device):
         for _, _, parameter in self.parameters():
             parameter.to(device)
+
+        return self
     
     def state_dict(self):
         state = OrderedDict()
diff --git a/norch/tensor.py b/norch/tensor.py
index 8063cba..e660df9 100644
--- a/norch/tensor.py
+++ b/norch/tensor.py
@@ -52,6 +52,8 @@ class Tensor:
                 self.ndim_ctype,
                 self.device_ctype
             )
+
+            self.to(device)
         
         else:
             self.tensor = None,
diff --git a/tests/test_autograd.py b/tests/test_autograd.py
index 6793816..823ba1b 100644
--- a/tests/test_autograd.py
+++ b/tests/test_autograd.py
@@ -24,11 +24,8 @@ class TestTensorAutograd(unittest.TestCase):
         norch_tensor1_grad = utils.to_torch(norch_tensor1.grad).to(self.device)
         norch_tensor2_grad = utils.to_torch(norch_tensor2.grad).to(self.device)
 
-        torch_tensor1 = torch.tensor([[[1, 2.5], [3, -4]], [[5, 6], [7, 8]]], requires_grad=True)
-        torch_tensor2 = torch.tensor([[[1, 1.], [1, 1.9]], [[1, 1], [1, 1]]], requires_grad=True)
-
-        torch_tensor1.to(self.device)
-        torch_tensor2.to(self.device)
+        torch_tensor1 = torch.tensor([[[1, 2.5], [3, -4]], [[5, 6], [7, 8]]], requires_grad=True, device=self.device)
+        torch_tensor2 = torch.tensor([[[1, 1.], [1, 1.9]], [[1, 1], [1, 1]]], requires_grad=True, device=self.device)
 
         torch_result = (torch_tensor1 + torch_tensor2).sum()
         torch_result.backward()