diff --git a/python/triton/code_gen.py b/python/triton/code_gen.py
index ae69f876c..70799861d 100644
--- a/python/triton/code_gen.py
+++ b/python/triton/code_gen.py
@@ -557,6 +557,14 @@ class Kernel:
         tensor_idxs = [i for i, arg in enumerate(wargs) if hasattr(arg, 'data_ptr')]
         if len(tensor_idxs) == 0:
             raise ValueError("No Tensor argument found.")
+        invalid_args = []
+        for idx in tensor_idxs:
+            curr = wargs[idx]
+            if not curr.is_cuda:
+                invalid_args += [idx]
+        if invalid_args:
+            raise ValueError("Arguments at index {invalid_args} are on the wrong device.".format(invalid_args=invalid_args) +
+                             " Only CUDA is supported at the moment")
         device = wargs[tensor_idxs[0]].device
         torch.cuda.set_device(device.index)
         # attributes